Tropospheric ozone variations governed by changes in stratospheric circulation

Thedownward transport of stratospheric ozone is an important natural source of tropospheric ozone, particularly in the upper troposphere, where changes in ozone have their largest radiative e ect1. Stratospheric circulation is projected to intensify over the coming century, which could lead to an increase in the flux of ozone from the stratosphere to the troposphere2–4. However, large uncertainties in the stratospheric contribution to trends and variability in tropospheric ozone levels5–7 make it di cult to reliably project future changes in tropospheric ozone8. Here, we use satellite measurements of stratospheric water vapour and tropospheric ozone levels collected between 2005 and 2010 to assess the e ect of changes in stratospheric circulation, driven by El Niño/Southern Oscillation and the stratospheric Quasi-Biennial Oscillation, on tropospheric ozone levels.Wefind that interannual variations in thestrength of the stratospheric circulation of around 40%—comparable to the mean change in stratospheric circulation projected this century2—lead to changes in tropospheric ozone levels in the northern mid-latitudes of around 2%, approximately half of the interannual variability. Assuming that the observed response of tropospheric ozone levels to interannual variations in circulation is a good predictor of its equilibrium response, we suggest that the projected intensification of the stratospheric circulation over the coming century could lead to small but important increases in tropospheric ozone levels. Modelling studies and observational analyses, the latter generally based on sparse data over limited regions, suggest that the stratospheremay play an important role inmodulating tropospheric ozone abundances5–7, but the magnitude of the stratospheric contribution and its importance in the tropospheric ozone budget are poorly constrained. Coupled chemistry–climatemodels (CCMs) consistently predict increases in the stratospheric circulation over the next century2–4, with corresponding circulation-driven increases in the stratosphere-to-troposphere (STE) ozone flux of 20–30% from 1965 to 20952. Most of these CCMs, however, lack comprehensive tropospheric chemistry and cannot be used to assess the impact of changes in the STE ozone flux on tropospheric ozone. A few tropospheric CCMs, which have poorly resolved stratospheres and often use prescribed stratospheric ozone, have examined the role of the stratosphere in future tropospheric ozone trends. They produce a much larger range of climate-driven STE ozone flux increase (25–80%) than stratospheric CCMs, and the magnitude of the change in STE ozone flux in some cases determines whether mid-latitude tropospheric ozone increases or decreases given other climateand emissions-driven changes in the budget9–11. Estimates of the tropospheric ozone response to changes in the STE ozone flux are thus critical to developing effective air quality policies. Six years of ozone and water vapour measurements from the Tropospheric Emission Spectrometer (TES) and Microwave Limb Sounder (MLS) onboard NASA’s Aura satellite (Supplementary Information 1) reveal that dynamical variability driven by El Niño/Southern Oscillation (ENSO) and/or the stratospheric Quasi-Biennial Oscillation (QBO) provides a ‘natural experiment’ that allows us to quantify the impact of changes in the stratospheric circulation on tropospheric ozone. As discussed below, ENSO and the QBO are highly correlated during the observational period, making it impossible to separate the contribution of the two modes to observed circulation changes. Previous studies have shown that both ENSO and theQBOmodulate the stratospheric circulation12–14 and ozone15–17, and, according to models, the STE ozone flux18–20. Other analyses have assessed variability in the stratospheric circulation and STE ozone flux using MLS measurements21,22, without examining the origins or tropospheric effects of this variability. This paper provides an observation-based end-to-end connection of these elements and an assessment of their impact on tropospheric ozone. As shown below and illustrated in Fig. 1, El Niño/easterly shear QBO (Fig. 1a,b) act to strengthen the stratospheric overturning circulation (particularly during winter, via mechanisms described in Supplementary Information 2) (Fig. 1c) and hence increase transport of air from the ozone maximum poleward and downward to mid-latitudes (Fig. 1d), which increases the STE ozone flux and thus tropospheric ozone (Fig. 1e). In contrast, La Niña/westerly shear QBO is associated with a weakened circulation and decreases in the STE flux and ozone. For the purpose of examining the tropospheric impact of changes in stratospheric transport, El Niño/easterly shear QBO therefore provides an analogue to climate change (with caveats discussed below) in which increased greenhouse gases (GHGs) are predicted to strengthen the circulation, leading to increased STE ozone flux2,3. Figure 2a shows the 2005–2010 time series of the Multivariate ENSO Index23 and aQBO shear index calculated from the difference in the Singapore zonal wind at 50 hPa and 25 hPa (U50−U25; ref. 24), along with deseasonalized zonal mean anomalies in stratospheric tropical upwelling derived from MLS water vapour measurements21 (Fig. 2b; Supplementary Information 3), northern mid-latitude lower stratospheric ozone from MLS (Fig. 2c), and northern mid-latitude mid-tropospheric ozone from TES (Fig. 2d). We focus on the Northern Hemisphere because variability in lower