A Study On Sensitivity Of Surface Ozone Formation To Its Precursors Based On Fractal Methods Of Time Series

In recent years,the control for particle pollution has achieved initial results,but the ozone（O₃）pollution in the atmosphere is becoming more and more serious.Surface O₃ is a secondary pollutant generated by primary pollutants such as nitrogen oxide（NO_x）and volatile organic compounds（VOC）after a series of photochemical reactions under solar radiation.The relationship between O₃ and its precursors is highly non-linear,and ozone-precursor relationship is generally split into a"NO_x-sensitive","VOC-sensitive"and"mixed sensitive".The ozone-precursor sensitive is of great guiding significance for formulating effective O₃ pollution control strategies and improving air quality.Due to the incomplete of emission inventory and insufficient understanding of photochemical mechanism,the numerical model based on the atmospheric chemical transport process still has great difficulties and uncertainties in the simulation process of ozone-precursor sensitive.In this study,based on the theory of complexity science and measured data,the fractal methods are introduced to study the ozone-precursor relationship from a macro and holistic perspective.In this paper,the study area is Taiwan Province of China,and a comparative study has been carried out on the typical areas in South and North Taiwan.In addition,since the monitoring index in Taiwan is non-methane hydrocarbon（NMHC）,which is a part of the total VOC.So,NMHC replaces VOCs as one precursor of ozone in this paper.The main content of this paper is mainly divided into four parts:（1）The power spectrum analysis method,detrended fluctuation analysis method（DFA）and multifractal method are used to test the fractal characteristic of O₃,NO_x and NMHC time series,which are from the traffic stations（Fengshan and Zhongli）and general stations（Nanzi and Zhongshan）in the South and North regions of Taiwan during2014-2018.（2）A new nonlinear sensitivity index（CSD and R）based on multifractal parameters is established by using the coupled detrended fluctuation analysis（CDFA）to quantify the Ozone-NO_x-NMHC sensitivity,and compared it with the traditional empirical dynamics method.Next,based on the indexes（CSD and R）and sliding window method,we further studied the temporal evolution process and the diurnal variation in summer and winter of Ozone-NO_x-NMHC sensitivity in 2018.（3）For Qiaotou and Chaozhou sites in southern Taiwan during 2013-2018,based on the CDFA method,a new index（PDB）was established to quantity the contribution of isoprene to O₃ formation;then the effects of different NO₂/NO ratios on the relationship between isoprene and surface ozone formation have been revealed futherly.（4）For Puzi site in central and Western Taiwan,for benzene and toluene which are the toxic components in NMHC,the index PDB is used to quantify the coupling correlation in Ozone-NO_x-benzene（toluene）system based on CDFA method.Furthermore,a comparative analysis of the correlation between PM_2.5 concentration and PDB value in daytime and nighttimethe in every winter from 2007 to 2016 was analyzed to reveal the environmental effect,which is whether PM_2.5 in the atmosphere promoted or inhibited the ozone photochemical system.The main results obtained in this paper are as follows:（1）Based on power spectrum analysis and DFA analysis,it is found that O₃,NO_x and NMHC time series have obvious long-term persistence of fractal features,which indicates that the concentration fluctuation is not independent in time scale.There is a certain"memory".Moreover,the three time series have obvious multifractal characteristics,and the multifractal degrees show regional difference in summer and winter.The regional and seasonal differences show that the complexity of O₃,NO_x and NMHC concentration fluctuations in different time scales and levels are different due to the combined effect of internal and external factors.（2）Based on the indexes（CSD and R）,the seasonal variation shows that,Ozone-NO_x-NMHC sensitivity is associated with“NMHC-sensitive”in high temperature seasons for Fengshan、Zhongli and Zhongshan sites;While in winter,Ozone-NO_x-NMHC sensitivity is almost associated with“mixed sensitive”for the three sites.However,Due to the difference of industrial layout,the Ozone-NO_x-NMHC sensitivity in Nanzi sites is obviously associated with"NO_x-sensitive"in summer and autumn,and associated with"mixed sensitive"in spring and winter.By comparing the results of EKMA curve and CDFA method,it is found that the results of the two methods are identical for Nanzi and Zhongshan sites,but some differences have been found between the two methods for the two traffic sites Fengshan and Zhongli.Compared with EKMA curve,CDFA method has many advantages,such as shorter time consumption,less uncertainty and stronger regional universality.（3）Based on the indexes（CSD and R）,the temporal evolution process of Ozone-NO_x-NMHC sensitivity on a high-precision time scale is studied.The results show that there are large regional differences in Taiwan.The differences of industrial distribution and meteorological factors between Southern and North Taiwan result in that the conversion process of Ozone-NO_x-NMHC sensitivity in Southern Taiwan is more frequent.Furthermore,the diurnal variation of Ozone-NO_x-NMHC sensitivity shows that the diurnal variation of Ozone-NO_x-NMHC sensitivity in summer and winter are different in Fengshan,Zhongshan and Nanzi.In summer,most of the time it is associated with“NMHC-sensitive”in Fengshan and Zhongshan;while in winter,most of the time it is associated with“NO_x-sensitive”.But Nanzi site is just the opposite of Fengshan and Zhongshan sites.In addition,the Ozone-NO_x-NMHC sensitivity in Zhongli sites presents a transition from"NMHC-sensitive"in the morning to"NO_x-sensitive"in the afternoon,which does not exist in other three sites.This transition may be related to the local pollution source,air mass aging,mixed layer height and horizontal transportation.（4）In recent years,the PDB values for Qiaotou site have a slight upward trend in spring and summer,but not obvious;while there is a downward trend of the PDB values in spring,summer and autumn for Chaozhou site.The correlation analysis of NO₂/NO ratio and PDB values in the summer from 2013 to 2018 shows that in the atmospheric environment with NO₂/NO<6,isoprene contributes a lot to the formation of surface ozone,and vice versa.（5）For Puzi sites located in Central and Western Taiwan,the PDB values of benzene and toluene were higher in the daytime than that in the nighttime during 2007～2016,and the difference of toluene in the daytime and nighttime was more significant.The correlation analysis of PM_2.5 concentration and PDB values in daytime and nighttime shows that,PDB values of benzene and toluene is positively correlated with PM_2.5concentration in daytime;while,they are almost irrelevant at nighttime.This shows that the surface heterogeneous reaction of PM_2.5 can promote the photolysis of benzene（toluene）,although PM_2.5 will inevitably scatter and block the solar radiation leading to weaken photochemical reactions.In summary,the fractal methods of time series can not only reveal Ozone-NO_x-NMHC sensitivity in a certain time scale,but also show its temporal dynamic process on a high-precision time scale and its transformation process in a small time scale.It is of great significance to understand the formation process and mechanism of O3.Moreover,compared with the traditional method,it is more simple and practicable.With the progress of air quality monitoring technology and the continuous improvement of observation system in China,the observation data of O₃and its precursors（especially VOC components）will be more abundant and reliable.Therefore,this method is expected to be widely used in the prevention and control of O₃ pollution in the future.