Identification Of Controlling Factors On The Long-term Trends Of Ozone Pollution Over Typical Areas Of China

Over recent years,air pollution in China,represented by fine particulate matter(PM_2.5) pollution,has been gradually improving.In contrast,surface-level ozone pollution is worsening.In the photochemically active areas such as Pearl River Delta（PRD）and Yangtze River Delta（YRD）,ozone has become the principal pollutant,contributing more polluted days than PM_2.5.Ozone pollution has become one of the bottleneck problems for air quality continuous improvement in China.Experiences from developed countries demonstrated that ozone pollution control is far more difficult than PM_2.5 pollution control.National Technical Guidelines for Ozone Pollution Source Apportionment clearly indicated that the difficulties of ozone pollution control lies in the following three aspects.First,ozone pollution can be transported in long distance,and ozone pollution control needs to elucidate the relative contributions from different areas.Second,the relationship between ozone and precursors is non-linear,and cutting precursor emissions need to follow certain ratios and be location-specific.Finally,emission sources of volatile organic compounds are complex,and their accurate control is difficult to achieve.Targeting the first two of the aforementioned three difficulties in ozone pollution control and in consideration of the lack in understand the characteristic the of regional ozone pollution,this study made the recognition of the key controlling factors of regional ozone pollution characteristic long-term trends,firstly,and then attempted to clarify the relative contributions of meteorology and precursor emissions from different areas and the non-linear responses of ozone and precursors,over the time scales of years to decades.By addressing the limitations of common research tools not appropriate for investigating long-term changes,we developed and optimized a series of novel research tools,including quantitative identification methods of meteorology and local and non-local emission changes to ozone pollution in combination of meteorological adjustment and source apportionment,and identification of long-term variations of ozone formation regimes based on two-dimensional multi-validation.We used these tools to investigate in detail the long-term variations of ozone pollution and its impacting factors over PRD and YRD.Identification of long-term trends of relative contributions of meteorology and local and non-local emission changes to ozone pollution and of ozone formation regimes facilitate us to gain more insight into the impacting mechanism of meteorological changes to ozone pollution in the context of climate changes and to accurately evaluate the potential impact of pollution control measures to ozone pollution over the past years.By drawing lessons from history,we would better formulate ozone pollution control strategies in both short-term and mid-to long-term over photochemically active areas of China,so as to curb and reverse the continuously deteriorating trend of ozone pollution,to achieve the goal of synergistic control of ozone and PM_2.5,and to ultimately protect public health.This Study has the following major findings and conclusions.（1）Spatial analysis technology was utilized to explore the aggregation of different stations in PRD.Global moran’s I（GM）,which manifest the aggregation and discrete distribution of stations overall was positive value,indicating that stations with similar ozone concentration in PRD were distributited aggregated.GM of annual averages showed an increasing from 2007 to 2017,especially from 2007 to 2012,demonstating that the discrepancies beween stations shrank in long-term scale.And further investigation implied local emission was the driving factor for the long-term variations of annual GM.Mothly GM was positive during ozone seasons,became negligible and even negative when ozone concentration was low,thus suggesting that ozone concentration in different stations approached to a similar level during ozone seasons and discrepancies in ozone concentration beween stations enlarged during non ozone seasons.Local moran’s I（LM）,which manifest the aggregation and discrete distribution of stations locally,distributed stations in PRD to the first and the third quadrants,revealing that stations in Hong Kong were in low ozone concentration and surrounded by low ozone levels stations,and stations in PRD except Hong Kong were in high ozone concentration and encircled by high ozone levels stations.（2）The correlation between ozone and meteorological factors showed a slight decrease during 2007 to 2018 in PRD,and that with surface net solar radiation（SSR） and temperature（T） inceaded during the periods.Conversely,correlation between ozone with relative humidity（RH） and wind decreaded.Detailly,the west of PRD belonged to meteorology-sensitive regions,T and SSR governed the patterns of spatial distributions of ozone with meteorology.Furthermore,ozone in south of PRD was negatively associated with wind and RH during the first few years,and the effect of RH became weak later.Dilution and transport of ozone were dominant to ozone by wind in south during the first and the last periods,and transport by wind was reinforced over time in north of PRD.（3）Ozone increase of PRD was suppressed by meteorological factors in long-term scale from 2007 to 2017 in general condition,while meteorology hardly made effects on long-trem trends during ozone episodes.Nonlocal contribution increased in both occasions.Differently,high nonlocal concentrated in central and southwestherly in general condition,while that concentrated in northestherly during episodes.The consistencies in time series of nonlocal in Beijing-Tianjin-Heibei（BTH）,East China（EC） and Guangdong（GD） with the averages in upwind indicated that there exist ozone transport between BTH and EC,EC and GD.（4）Impacts from local emission on different stations within PRD were disparate in general condition and during ozone episodes owing to the different ozone accumulation and formation mechanism in PRD.Local contribution was positive in northeast and negative in southwest,owing to the promotion of NO_x from central to zone formation for the northeast in NO_x-limited,and the consumption of NO_x to ozone for the southwest in VOC-limited in general condition.The contribution in northeast decreased and that in northeast increased in southwest with the priority in NO_x reductions.Conversely during ozone episodes,local contribution was negative in northeast and positive in southwest,owing to the export of ozone from northeast and the sedimentation of ozone in southwest.Their long-term trends decreased in northeast and southwest where belonged to NO_x-limited as NO_x decresead,and increased in central PRD where was VOC-limited.（5）Ozone formation mechanism recognized from NO_x and temperature kept highly coincident in both Shanghai and PRD based on the two dimensional recognition and validation model.Overall,the west of Shanghai was more sensitive to the organic reactivity（VOCR）where were affected by large NO_x emissions from Jiangsu and Zhejiang,and the east was more NO_x-sensitive.Sensitivities changed little in west and the east became more NO_x-sensitive from 2008 to 2019.Contrasting that with Shanghai,the central of PRD was tended to transition region and the periphery was more NO_x-sensitive.The west became more NO_x-sensitive and the sensitivity to NO_x regions in east was shank to transition region from 2013 to 2018.The significant difference in ozone formation mechanism evolution between Shanghai and PRD is that the mechanism in the west of Shanghai changed little owing to the affects by large NO_x emission from Jiangsu and Zhejiang,while that in the rest of Shanghai of PRD changed a lot as the emission reduction.