Temperature dependence of tropospheric ozone under NOx reductions over Germany

Abstract

Due to the strong temperature dependence of surface ozone concentrations (O3), future warmer conditions may worsen ozone pollution levels despite continued efforts on emission controls of ozone precursors. Using long-term measurements of hourly O3 concentrations co-located with NOx concentrations in stations distributed throughout Germany, we assess changes in the climate penalty in summertime, defined as the slope of ozone-temperature relationship during the period 1999–2018. We find a stronger temperature sensitivity in the urban stations over the southwestern regions, especially in the first period of the study (1999–2008). We show a decrease in the climate penalty in most of stations during the second period of the study (2009–2018), with some exceptions (e.g. Berlin) where the climate penalty did not show significant changes. A key motivation of this study is to provide further insights into the impacts of NOx reductions in the O3-temperature relationship. For that, we propose a statisti-cal approach based on Generalized Additive Models (GAMs) to describe ozone production rates, inferred from hourly observations, as a function of NOx and temperature, among other variables relevant during the O3 production. The GAMs confirm lower O3 production rates during the second period (2009–2018) at most of the stations and a decreasing sensitivity to temperature. We observe that a large number of stations are transitioning to NOx-limited chemistry, consistent with a decreasing temperature dependence of O3 at moderate-high temperatures as a result of sustained NOx reductions. Moreover, the GAMs results showed changes in the shape of the function representing the O3-temperature relationship when comparing the first and second period, which suggest changes in VOC influencing the temperature dependence of O3. From these results, we infer effective VOC reductions over time that have also contributed to the observed decrease of O3 production rates. Thus, our analysis indicates that emissions reductions have been effective in a number of stations, particularly in the southwestern regions. However, we notice that in a few stations (e.g. Berlin) additional emission reductions should be required to effectively mitigate the temperature dependence of O3.