First long-term surface ozone variations at an agricultural site in the North China Plain: Evolution under changing meteorology and emissions.

Significant upward trends in surface ozone (O) have been widely reported in China during recent years, especially during warm seasons in the North China Plain (NCP), exerting adverse environmental effects on human health and agriculture. Quantifying long-term O variations and their attributions helps to understand the causes of regional O pollution and to formulate according control strategy. In this study, we present long-term trends of O in the warm seasons (April-September) during 2006-2019 at an agricultural site in the NCP and investigate the relative contributions of meteorological and anthropogenic factors. Overall, the maximum daily 8-h average (MDA8) O exhibited a weak decreasing trend with large interannual variability. < 6 % of the observed trend could be explained by changes in meteorological conditions, while the remaining 94 % was attributed to anthropogenic impacts. However, the interannual variability of warm season MDA8 O was driven by both meteorology (36 ± 28 %) and anthropogenic factors (64 ± 27 %). Daily maximum temperature was the most essential factor affecting O variations, followed by ultraviolet radiation b (UVB) and boundary layer height (BLH), with rising temperature trends inducing O inclines throughout April to August, while UVB mainly influenced O during summer months. Under changes in emissions and air quality, warm season O production regime gradually shifted from dominantly VOCs-limited during 2006-2015 to NO-limited afterwards. Relatively steady HCHO and remarkably rising NO levels resulted in the fast decreasing MDA8 O (-2.87 ppb yr) during 2006-2012. Rapidly decreasing NO, flat or slightly increasing HCHO promoted O increases during 2012-2015 (9.76 ppb yr). While afterwards, slow increases in HCHO and downwards fluctuating NO led to decreases in MDA8 O (-4.97 ppb yr). Additionally, continuous warming trends might promote natural emissions of O precursors and magnify their impacts on agricultural O by inducing high variability, which would require even more anthropogenic reduction to compensate for.