Interactions between clouds and atmospheric circulation in the extratropics

Abstract

In climate models, the simulation of clouds is known to be particularly problematic, leading to important biases in the climatological energy balance on regional scales, as well as to large uncertainties in the future amount of warming in response to greenhouse gas increase. This thesis explores the connections between clouds and atmospheric circulation in extratropical regions. In particular, we investigate the impacts of clouds and their uncertainties on atmospheric circulation and its response to global warming. We find that clouds have very substantial effects both on the mean circulation and on its future response to warming in climate models. In the mean state, the position of the midlatitude jet correlates well with the midlatitude shortwave cloud-radiative effect (SW CRE), which suffers from very large biases in models. Models in which midlatitude SW CRE is too negative have anomalously cold midlatitudes, leading to an anomalously equatorward jet position. This result is supported by idealized model experiments and appears consistent with the effect of midlatitude baroclinicity changes on eddy activity. This means that an accurate representation of clouds and their radiative effects is essential to correctly portray the mean circulation. In the context of greenhouse gas–forced change, we demonstrate that cloud-radiative changes have a surprisingly large impact on the atmospheric circulation response. This results mainly from the SW cloud feedback, whose specific spatial structure induces low-latitude warming and high-latitude cooling, enhancing midlatitude baroclinicity and favoring a strengthening and poleward shift of the midlatitude jet. This opposes the effects of other major feedbacks (e.g., the water vapor feedback and the longwave cloud feedback), which produce polar-amplified warming and weakened midlatitude baroclinicity. For this reason, cloud-radiative changes explain the majority of the poleward expansion of atmospheric circulation in our model, even though the net cloud feedback only explains one-fourth of the total warming in our experiment. The importance of clouds for the forced atmospheric circulation response is confirmed by considering global warming model experiments of the CMIP5 archive. We demonstrate that the Southern Hemispheric jet stream response is strongly linked to the meridional structure of the SW feedbacks around the midlatitudes, which are primarily driven by clouds and sea ice. Consistent with our previous findings, models in which the gradient of absorbed shortwave radiation increases (e.g., midlatitude baroclinicity is enhanced) tend to yield a larger poleward shift of the midlatitude jet. An important implication of this result is that uncertainties in the cloud feedback produce uncertainties in the atmospheric circulation response. This provides additional motivation to reduce the inter-model spread in cloud feedbacks, not only in terms of their impact on the global energy balance, but also in terms of their spatial structure.