Energy transport, polar amplification, and ITCZ shifts in the GeoMIP G1 ensemble

Abstract. The polar amplification of warming and the ability of the Intertropical Convergence Zone (ITCZ) to shift to the north or south are two very important problems in climate science. Examining these behaviors in global climate models (GCMs) running solar geoengineering experiments is helpful not only for predicting the effects of solar geoengineering but also for understanding how these processes work under increased carbon dioxide (CO₂). Both polar amplification and ITCZ shifts are closely related to the meridional transport of moist static energy (MSE) by the atmosphere. This study examines changes in MSE transport in 10 fully coupled GCMs in experiment G1 of the Geoengineering Model Intercomparison Project (GeoMIP), in which the solar constant is reduced to compensate for the radiative forcing from abruptly quadrupled CO₂ concentrations. In G1, poleward MSE transport decreases relative to preindustrial conditions in all models, in contrast to the Coupled Model Intercomparison Project phase 5 (CMIP5) abrupt4xCO2 experiment, in which poleward MSE transport increases. We show that since poleward energy transport decreases rather than increases, and local feedbacks cannot change the sign of an initial temperature change, the residual polar amplification in the G1 experiment must be due to the net positive forcing in the polar regions and net negative forcing in the tropics, which arise from the different spatial patterns of the simultaneously imposed solar and CO₂ forcings. However, the reduction in poleward energy transport likely plays a role in limiting the polar warming in G1. An attribution study with a moist energy balance model shows that cloud feedbacks are the largest source of uncertainty regarding changes in poleward energy transport in midlatitudes in G1, as well as for changes in cross-equatorial energy transport, which are anticorrelated with ITCZ shifts.