Multi-factor reconciliation of discrepancies in ozone-precursor sensitivity retrieved from observation- and emission-based models

Abstract

Ground-level O3 pollution has been continuously worsening in China despite gradual improvement in other major pollutant levels. Understanding the sensitivity of O3 production to its precursors (OPS) is a prerequisite for formulating effective O3 control measures, but this has been hampered by significant discrepancies in OPS produced by traditional identification approaches using observation-based models (OBM) and emission-based models (EBM). In this study, by applying OBM and EBM in parallel within a month having significant O3 pollution in Shanghai, China, we demonstrated that a lack of carbonyl input, overestimation in NO2 monitoring data, and differences in simulation period and emission reduction area were the core factors leading to OPS discrepancies, and that a reliable OPS cannot be obtained unless these factors are reconciled. By collectively addressing these factors, the number of days with a consistent OPS from both models increased from 6–7 to 20–21 in a month, and the R value defined to quantify the discrepancy decreased by ∼55%. The contributions of these factors to OPS discrepancy differed greatly in urban and suburban settings, mainly caused by differences in pollutant emission and transport characteristics. Overall, OPS identified solely by OBM or EBM is associated with great uncertainty, while reliable OPS estimation can be achieved by a collective application of OBM and EBM with consensus on the above factors. The method demonstrated here could be applied to other photo-chemically active regions worldwide as part of efforts to address ozone pollution.

Keywords

• Ozone
• Observation-based model
• Emission-based model
• O3 precursor sensitivity
• Discrepancy reconciliation