Sensitivity analysis and precursor emission sources reduction strategies of O 3 for different pollution weather types based on the GRAPES-CUACE adjoint model

In recent years, China has continued to face persistent high concentrations of atmospheric O 3 . While various methods can analyze the non-linear relationship between precursor emissions and O 3 concentrations, challenges remain in accurately quantifying the contributions of specific emission sources to O 3 levels due to the complexity of atmospheric chemistry and varying meteorological conditions. At the same time, changes in the atmospheric circulation situation can also affect the distribution of sensitive source areas for O 3 pollution in a given area. Therefore, the study of O 3 '' source-concentration " relationship under different types of polluted weather is of great significance for the development of scientific and effective O 3 control programmes. This study firstly screened the O 3 pollution processes based on the observed O 3 concentration data in Beijing, and summarised the O 3 pollution processes in Beijing into four types using subjective typing methods :Ridge-Internal Type (Type 1), Westward Flow Type (Type 2), Weak Ridge with Northwest Flow Type (Type 3), and Southwest Flow ahead of Trough Type (Type 4). The GRAPES-CUACE adjoint model version 2.0 was then used to carry out adjoint sensitivity analyses of O 3 concentrations for each type of individual case, to quantitatively assess the contribution of local and neighbouring precursor emission sources to O 3 concentrations in Beijing, and to further explore O 3 control options for each type of individual case. Adjoint sensitivity analyses showed that the four types of O 3 generation were controlled by different precursor emission sources and that there were significant differences in the distribution of sensitive source areas. Among the different types of individual cases, there were slight differences in the contributions of precursor emission sources from Beijing, Tianjin, Hebei, Shanxi and Shandong to the Maximum Daily 8 -Hour Average O 3 concentrations (MDA8 O 3 ) in Beijing. In general, Beijing and Hebei sources contributed the most to the maximum 8-h average concentration of O 3 in Beijing, with a cumulative contribution of 66.4 -85.6%. The emission reduction results of the 4 types of cases showed that the MDA8 O 3 in Beijing can be made to reach the set standard (70 ppb) through several iterations, which reflected the high efficiency of the adjoint model in the design of emission reduction experiments. When the O 3 concentrations of the four types of individual cases in this study were reached, the reduction percentages of NO x , VOCs and CO were 49.2 -83.0%, 40.6 -65.3% and 34.7 -67.6%, respectively, in each province and city. By identifying the key precursor emission sources and their contributions to ozone levels under different weather types, this study provides good guidance for policymakers and environmental managers to develop more effective ozone reduction strategies. Specifically, the ability to predict the impact of various emission reduction scenarios under different weather conditions can help in designing targeted interventions that are both efficient and costeffective.