Long-term change in the upper atmosphere and a possible effect on surface climate

Not only the climate in the troposphere is changing, also at higher altitudes long-term trends have been observed. The global mean middle and upper atmosphere have been cooling, resulting in atmospheric contraction and a decline in thermosphere density at fixed height, mainly driven by the increase in atmospheric CO₂ concentration. The secular variation of the Earth’s magnetic field is an additional driver of long-term change in the upper atmosphere, causing trends that are strongly location-dependent. While magnetic field changes are most important for the ionosphere, they also affect the temperature and wind structure throughout the thermosphere, mainly via changes in the strength and geographic distribution of Joule heating. Simulations with the Whole Atmosphere Community Climate Model eXtension (WACCM-X) suggest that perturbations induced by magnetic field changes in the lower thermosphere climate can further propagate downward via vertical dynamical coupling. Our results show a significant response in the zonal mean temperature and zonal wind in the Southern Hemisphere (SH) middle- to high-latitude troposphere, stratosphere, and mesosphere of up to ±2 K and ±2 m/s, as well as regionally significant changes in Northern Hemisphere (NH) polar surface temperatures of up to ±1.3 K, in December-January-February. In the SH, changes in gravity wave filtering in the thermosphere induce a change in the residual circulation that extends down into the upper mesosphere, where further changes in the zonal mean wind climatology are generated, together with changes in local planetary wave generation and/or amplification and gravity wave filtering. This induces an anomalous residual circulation that extends down into the troposphere. The NH middle atmosphere response is zonally asymmetric, consisting of a significant change in the positioning and shape of the upper stratospheric polar vortex, which is dynamically consistent with the surface temperature response. While the details of the lower and middle atmosphere responses may not be entirely accurate due to model limitations, the results from our simulations do indicate that dynamical coupling within the atmosphere can conceivably result in upper atmosphere processes having a significant effect on surface climate.