Dynamical-Chemical Feedbacks in General Circulation Models and Their Influence on Sudden Stratospheric Warming Events

Sudden Stratospheric Warming events (SSWs) are rapid disruptions of the Northern Hemisphere (NH) winter stratospheric polar vortex and represent the largest source of inter-annual variability in the NH winter stratosphere. They have been linked to winter surface climate anomalies such as cold snaps over North America and Eurasia. Representing these events accurately in large scale GCMs as well as developing a greater understanding of them is key to improving predictability of winter surface climate. A key component of a GCM is its representation of atmospheric chemistry. Chemical distributions are either prescribed or calculated interactively by coupling an atmospheric chemistry model to radiation and dynamical components, thus capturing any chemical dynamical feedback mechanisms but incurring significant running cost.

This work evaluates the impact of interactive chemistry when modelling SSW events and explores the feedback mechanisms between chemical distributions and stratospheric dynamical variability. Pre-industrial control runs from the MetOffice HadGEMGC3.1 model which prescribes chemical fields and UKESM1 which calculates trace gas concentration interactively are utilised. Over the whole season - The Earth System Model appears to suppress warmings while the model with prescribed physics overestimates their occurrence compared to reanalysis. The differing representation of the equatorial stratosphere appears to be partially responsible for this difference. Additionally we find that middle stratosphere equatorial ozone concentration in late NH summer is closely associated with SSW probability in the ensuing winter in UKESM1. Anomalously low ozone is generally associated with an elevated SSW rate. This implies a chemical-dynamical coupling between the equator and the vortex in this model which preliminary results suggest could be driven by chemical feedbacks influencing the state of the early winter Quasi Biennial Oscillation (QBO) and Semi-Annual Oscillation (SAO) in zonal winds which can alter the distribution of planetary wave propagation and breaking (the primary cause of SSWs). Further work will assess whether this phenomenon is observed in other GCMs and further explore the physical mechanisms responsible.