Improvement And Application Of Source Contribution Analysis Method Of PM_2.5 Based On Response Surface Modeling Technique

PM_2.5 is one of the major ecological and environmental problems in building a moderately prosperous society in all respects in China.PM_2.5 is affected by both primary PM_2.5 emissions and precursor emissions,which are produced from both local emissions and regional transport.There is a simple linear relationship between PM_2.5 and primary PM_2.5 emissions but a strong nonlinear response relationship between PM_2.5 and precursor emissions.Therefore,accurately identifying and quantifying source contributions of pollutant emissions to PM_2.5 concentration is a prerequisite for making an effective control strategy to alleviate PM_2.5 pollution and break through the bottleneck in reducing PM_2.5 level over the Pearl River Delta（PRD）region.In this paper,the source contribution analysis of PM_2.5 based on response surface model（RSM）technology was carried out.First,the sectoral linearity technique was newly developed and coupled in the latest RSM to build the nonlinear response relationship of PM_2.5 to precursor emissions.Meanwhile the linear response relationship of PM_2.5 to primary PM_2.5 emissions was integrated to develop a new RSM system.In addition,an innovative differential method（DM）based on RSM was proposed.The new DM,with the ability to reproduce the nonlinear response surface of PM_2.5 to precursor emissions by dissecting the emission changes into a series of small increments,has shown to overcome the issue of the traditional brute force method（BFM）in apportioning the source contribution of precursor emissions to PM_2.5.Finally,applying the source contribution analysis method based on RSM technology,this study investigated the contributions of emissions from multiple regions,sectors and pollutants to the PM_2.5 concentration in receptors over the PRD region in 2015.The results showed that the predicted PM_2.5 concentration of the optimized RSM had high consistency with the simulated one of CMAQ in the corresponding scenario.The average MNEs and MFEs were less than 2%;and the average Rs were greater than 0.999.The innovative DM based on RSM can effectively solve the problem of overestimation of the accumulative contribution of precursor emissions to PM_2.5 analyzed by BFM.Among all pollutants,PM_2.5 was most sensitive to primary PM_2.5 emissions;among all precursors,PM_2.5 was more sensitive to NH3 emissions.With the emission reduction ratio increasing,the sensitivity of PM_2.5 to NOX emissions would increase.The PM_2.5 levels over the PRD region was generally dominated by local emission sources,about 39-64%.Among the contributions of PM_2.5 from various sectors and pollutants,the primary PM_2.5 emissions from fugitive dust source contributed most to PM_2.5 levels,accounting for 25-42%.The contributions of agriculture NH3 emissions（6-13%）could also play a significant role compared to other sectoral precursor emissions.Among the NOX sectors,the emissions control of stationary combustion source could be most effective in reducing PM_2.5 levels over the PRD region.