Sensitivity of modelled urban background ozone concentrations to uncertainties in the GRS input variables

MAZZEO, N.A.; PINEDA ROJAS, A.L.

Budapest

Conferencia; 17th International Conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes (HARMO17); 2016

In this work, we apply the Monte Carlo analysis to evaluate the uncertainty of modelled summer maximum ozone diurnal peak concentrations (Cmax) in the Metropolitan Area of Buenos Aires (MABA), Argentina resulting from uncertainties in the Generic Reaction Set (GRS) input variables, using the DAUMOD-GRS model. Values of Cmax occurring at early morning or late evening hours present greater uncertainties than those occurring around midday hours. Uncertainty of Cmax is dominated by that in the GRS ozone initial concentration at all analysed receptors, with relative  contributions varying between 67.5-89.8%. The second most important variable is the nitrogen oxides initial concentration, whose relative contribution may increase (in our experiments up to 11.7%) depending on the uncertainties of the GRS input variables.