Five Decades of Baseline Ozone Trends at Northern Midlatitudes: Reconciling Observations and Models

Recent community-wide reviews highlight that there remains a challenge for explaining tropospheric O₃ trends over the past half century despite decades of research. Many prior studies show that models have difficulty simulating O₃ increases measured at northern mid-latitude remote sites, raising concerns about the utility of models for estimating tropospheric O₃ radiative forcing and assessing pollution control strategies. Here we use a suite of multi-decadal model hindcasts (1960-2015), in-situ and satellite observations to show that we can reconcile observed and simulated O₃ trends with internal climate variability, measurement sampling biases, and improved model representation of remote baseline conditions. Satellite measurements of mid-tropospheric O₃ and CO show that trans-Pacific pollution transport strengthens during strong El Niño events and weakens during La Niña. In 1998-1999, the Pacific Decadal Oscillation changed to the positive phase, tending towards more frequent La Niña events, leading to a decrease in transport from Asia towards the U.S. West Coast in the 2000s. Springtime surface O₃ measured at Lassen California (~2 km altitude) increased rapidly in the 1990s but levelled off in the 2000s, reflecting the interplay between the decrease in transport and an increase in Asian O₃ precursor emissions. The tripling of Asian NO_x emissions since 1990 contributes as much as 65% of observed springtime O₃ increases at western U.S. rural sites, outpacing the O₃ decreases attained via 50% US NO_x emission controls. Over Europe, stratospheric influence contributes to decadal trends (0.2 ppb yr^-1) of surface O₃ measured at Mace Head during 1990-2015, but cannot explain the observed wintertime O₃ increase at Zugspitze in the 1980s (0.7 ppb yr^-1). We show that a rapid switch to the positive phase of the North Atlantic Oscillation in the 1980s strengthens baseline airflow from the North Atlantic Ocean to Europe, which helps in ventilating European boundary layer air, leading to a decrease in surface O₃ removal by NO_x titration during winter. Over the 2004-2015 OMI/MLS satellite era, changes in anthropogenic emissions and meteorological variability contribute equally to the increases in global tropospheric O₃ burden. Our findings highlight that attribution of the decadal trends of tropospheric O₃ derived from satellite and in situ records requires consideration of internal climate variability.