Modeling Europe with CAMx for the Air Quality Model Evaluation International Initiative (AQMEII)
Abstract
The CAMx photochemical grid model was used to model ozone (O3) and particulate matter (PM) over a European modeling domain for calendar year 2006 as part of the Air Quality Model Evaluation International Initiative (AQMEII). The CAMx base case utilized input data provided by AQMEII for emissions, meteorology and boundary conditions. Sensitivity of model outputs to input data was investigated by using alternate input data and changing other important modeling assumptions including the schemes to represent photochemistry, dry deposition and vertical mixing. Impacts on model performance were evaluated by comparisons with ambient monitoring data. Base case model performance for January and July 2006 exhibited under-estimation trends for all pollutants both in winter and summer, except for SO2. SO2 generally had little bias although some over-estimation occurred at coastal locations and this was attributed to incorrect vertical distribution of emissions from marine vessels. Performance for NOx and NO2 was better in winter than summer. The tendency to under-predict daytime NOx and O3 in summer may result from insufficient NOx emissions or overstated daytime dilution (e.g., too deep planetary boundary layer) or monitors that are located near sources (e.g., roadside monitors). Winter O3 was biased low and this was attributed to a low bias in the O3 boundary conditions. PM10 was widely under-predicted in both winter and summer. The poor PM10 was influenced by underestimation of coarse PM emissions. Sensitivities of O3 concentrations to precursor emissions are quantified using the decoupled direct method in CAMx. The results suggest that O3 production over the central and southern Europe during summer is mostly NOx-limited but for a more northerly city, London, O3 production can be limited either by NOx or VOC depending upon daily meteorological conditions.