A Sensitivity Analysis of Ozone Formation to Ambient Air Composition by Means of Photochemical Models

The prediction of the atmospheric concentration of ozoneand of other photochemical pollutants represents one of themajor scientific challenges in the study of urban airquality. To this aim, photochemical models are widelyapplied as support tools for regulatory purposes, but theirresults are known to be affected by several uncertainties,in boundary conditions, geographical and meteorologicaldata and the chemical composition of emission and ambientair as well. Also the mechanism selected to describe theurban chemistry may lead to large differences in the modelresults. In this work, the influence of ambient air composition onthe results of photochemical model has been studied using a simple Lagrangian trajectory model coupled with different chemical sub-models (CB-IV and a simplified version of EMEP). To this aim a parametric sensitivity study of ozone with respect to its precursors' concentration was performed withreference to the same meteorological conditions and by simulatingdifferent scenarios characterized by specific emitted pollutantsconcentration. Rather than analyzing sensitivity by changing initial composition, an analytical local sensitivity concept has been applied; this allowed evaluation the influence of initial concentration on the predicted ozone concentration andto discriminate between VOC-sensitive and NOx-sensitive kineticregimes. Finally, these results have been successfully comparedwith those obtained applying the concept of photochemical indicators (i.e. O₃/NOy and H₂O₂/HNO₃ ratios).