The Steady-State Atmospheric Circulation Response to Climate Change–like Thermal Forcings in a Simple General Circulation Model

Abstract

The steady-state extratropical atmospheric response to thermal forcing is investigated in a simple atmospheric general circulation model. The thermal forcings qualitatively mimic three key aspects of anthropogenic climate change: warming in the tropical troposphere, cooling in the polar stratosphere, and warming at the polar surface. The principal novel findings are the following: 1) Warming in the tropical troposphere drives two robust responses in the model extratropical circulation: poleward shifts in the extratropical tropospheric storm tracks and a weakened stratospheric Brewer–Dobson circulation. The former result suggests heating in the tropical troposphere plays a fundamental role in the poleward contraction of the storm tracks found in Intergovernmental Panel on Climate Change (IPCC)-class climate change simulations; the latter result is in the opposite sense of the trends in the Brewer–Dobson circulation found in most previous climate change experiments. 2) Cooling in the polar stratosphere also drives a poleward shift in the extratropical storm tracks. The tropospheric response is largely consistent with that found in previous studies, but it is shown to be very sensitive to the level and depth of the forcing. In the stratosphere, the Brewer–Dobson circulation weakens at midlatitudes, but it strengthens at high latitudes because of anomalously poleward heat fluxes on the flank of the polar vortex. 3) Warming at the polar surface drives an equatorward shift of the storm tracks. The storm-track response to polar warming is in the opposite sense of the response to tropical tropospheric heating; hence large warming over the Arctic may act to attenuate the response of the Northern Hemisphere storm track to tropical heating. 4) The signs of the tropospheric and stratospheric responses to all thermal forcings considered here are robust to seasonal changes in the basic state, but the amplitude and details of the responses exhibit noticeable differences between equinoctial and wintertime conditions. Additionally, the responses exhibit marked nonlinearity in the sense that the response to multiple thermal forcings applied simultaneously is quantitatively different from the sum of the responses to the same forcings applied independently. Thus the response of the model to a given thermal forcing is demonstrably dependent on the other thermal forcings applied to the model.