LAUSR

Abstract Southern China has experienced severe photochemical pollution events in recent years, and the tropospheric ozone has emerged as the major pollutant of concern. Despite some recent efforts, the role of natural factors in constraining long-term trends of ozone in this region is poorly understood. In this study, we addressed this issue using tropospheric column ozone (TCO) datasets (2005–2017) from the Ozone Monitoring Instrument/Microwave Limb Sounder (OMI/MLS) and surface ozone datasets from 16 monitoring stations in Southern China (2006–2016). Consequently, we studied the influence of atmospheric dynamical factors, such as solar cycle, El-Nino Southern Oscillation (ENSO), Quasi-Biennial Oscillation (QBO), and Arctic Oscillation (AO), and local-scale meteorological factors, such as precipitation, surface temperature, planetary boundary layer height, and horizontal winds, on regional ozone trends. Stepwise multivariate regression analysis and harmonic function fitting were adopted to quantitatively simulate the influence of these natural drivers on ozone change. We found that within our research periods, both surface ozone and TCO in Southern China had a significant upward trend, with slopes of 0.97% y−1 (0.23 ppbv y−1) and 0.82% y−1 (0.28 DU y−1), respectively. Natural factors explained 44.4% of the TCO uptrend and 27.0% of the surface ozone uptrend. Among the natural factors, the solar cycle plays the most important role in tropospheric and surface ozone modulation. Its 11-year cycle had a large impact on TCO for 2–7 DU and on surface ozone for 3–8 ppbv. However, the ENSO, QBO, and AO indices did not affect tropospheric ozone trends significantly. In addition, we showed that precipitation and wind fields associated with the Asian summer monsoon played a critical role in lowering ozone levels over Southern China, accounting for 24.8% and 81.5% of summertime TCO and surface ozone variability, respectively. Finally, a significant fraction of TCO and surface ozone uptrends (55.6% and 73.0%, respectively) remained unexplained even after consideration of these natural factors in the periods 2005–2017 and 2006–2016, respectively. These unexplained factors are most likely related to anthropogenic emissions and should be studied further.