Stratospheric impact on the Northern Hemisphere winter and spring ozone interannual variability in the troposphere

Abstract

Abstract. In this study we use O3 and stratospheric O3 tracer simulations from the high-resolution Goddard Earth Observing System, Version 5 (GEOS-5) Replay run (MERRA-2 GMI at 0.5° model resolution ~ 50 km) and observations from ozonesondes to investigate the interannual variation and vertical extent of the stratospheric ozone impact on tropospheric ozone. Our work focuses on the winter and spring seasons over North America and Europe. The model reproduces the observed interannual variation of tropospheric O3, except for the Pinatubo period from 1991 to 1995 over the region of North America. Ozonesonde data show a negative ozone anomaly in 1992–1994 following the Pinatubo eruption, with recovery thereafter. The simulated anomaly is only half the magnitude of that observed. Our analysis suggests that the simulated Stratosphere-troposphere exchange (STE) flux deduced from the analysis might be too strong over the North American (50° N–70° N) region after the Mt. Pinatubo eruption in the early 1990s, masking the impact of lower stratospheric O3 concentration on tropospheric O3. European ozonesonde measurements show a similar but weaker O3 depletion after the Mt. Pinatubo eruption, which is fully reproduced by the model. Analysis based on a stratospheric O3 tracer (StratO3) identifies differences in strength and vertical extent of stratospheric ozone influence on the tropospheric ozone interannual variation (IAV) between North America and Europe. Over North America, the StratO3 IAV has a significant impact on tropospheric O3 from the upper to lower troposphere and explains about 60 % and 66 % of simulated O3 IAV at 400 hPa, ~ 11 % and 34 % at 700 hPa in winter and spring respectively. Over Europe, the influence is limited to the middle to upper troposphere, and becomes much smaller at 700 hPa. The stronger and deeper stratospheric contributions in the tropospheric O3 IAV over North America shown by the model is likely related to ozonesondes' being closer to the polar vortex in winter with lower geopotential height, lower tropopause height, and stronger coupling to the Arctic Oscillation in the lower troposphere (LT) than over Europe.