Impact of policy-relevant scenarios on ozone in southern England: Influence of chemical mechanism choice

Monte Carlo sampling of pre-specified parameter ranges has been used to replace single ‘best estimate’ Photochemical Trajectory Model runs with 9006 ‘acceptable’ model runs that were each consistent with the observations of elevated O3 during July 2006 at a rural location in southern England. These acceptable parameter sets were then used for probabilistic evaluation of policy relevant photochemical oxidant control strategies, based on 30% reductions in VOC and NOx precursor emissions and the geographic locations where the emission reductions were carried out. Indicator species ratios were used to complete a diagnostic evaluation of model performance. The chemical mechanism employed in the base case model was replaced with four other mechanisms from the literature. Probabilistic uncertainty analysis showed that the results of the policy-relevant scenarios showed little or no sensitivity to chemical mechanism choice. A large degree of error compensation could be achieved through the selection of acceptable parameter sets and this hid the differences between the five chemical mechanisms studied.

► Monte Carlo techniques can quantify uncertainties in ozone predictions.
► VOC- versus NOx-sensitivity changes from day-to-day.
► Impacts of policy-relevant scenarios are independent of chemical mechanisms.