Isotopic evidence of multiple controls on atmospheric oxidants over climate transitions

The abundance of tropospheric oxidants, such as ozone (O-3) and hydroxyl (OH) and peroxy radicals (HO2 + RO2), determines the lifetimes of reduced trace gases such as methane and the production of particulate matter important for climate and human health. The response of tropospheric oxidants to climate change is poorly constrained owing to large uncertainties in the degree to which processes that influence oxidants may change with climate(1) and owing to a lack of palaeo-records with which to constrain levels of atmospheric oxidants during past climate transitions(2). At present, it is thought that temperature-dependent emissions of tropospheric O-3 precursors and water vapour abundance determine the climate response of oxidants, resulting in lower tropospheric O-3 in cold climates while HOx (= OH + HO2 + RO2) remains relatively buffered(3). Here we report observations of oxygen-17 excess of nitrate (a proxy for the relative abundance of atmospheric O-3 and HOx) from a Greenland ice core over the most recent glacial-interglacial cycle and for two Dansgaard-Oeschger events. We find that tropospheric oxidants are sensitive to climate change with an increase in the O-3/HOx ratio in cold climates, the opposite of current expectations. We hypothesize that the observed increase in O-3/HOx in cold climates is driven by enhanced stratosphere-to-troposphere transport of O-3, and that reactive halogen chemistry is also enhanced in cold climates. Reactive halogens influence the oxidative capacity of the troposphere directly as oxidants themselves and indirectly(4) via their influence on O-3 and HOx. The strength of stratosphere-to-troposphere transport is largely controlled by the Brewer-Dobson circulation(5), which may be enhanced in colder climates owing to a stronger meridional gradient of sea surface temperatures(6), with implications for the response of tropospheric oxidants(7) and stratospheric thermal and mass balance(8). These two processes may represent important, yet relatively unexplored, climate feedback mechanisms during major climate transitions.