Quantification of the urban air pollution increment and its dependency on the use of down-scaled and bottom-up city emission inventories

The impact of large cities on air pollution levels usually is determined with models driven by so-called downscaled emission inventories. This implies that annual emissions of air pollutants at the national scale are spatially distributed over a grid using proxy data like population density. These inventories are frequently used for regional air quality modeling but increasingly used also to assess air quality over cities. A key question is how much the assessment of city air pollution and exposure is influenced by downscaling of emission input data. Within this study we compare modeled nitrogen dioxide and particulate matter (PM) concentrations driven by a downscaled emission inventory with modeled concentrations driven by a 'bottom-up' emission inventory for the Paris region. Predicted concentrations and gradients in the concentrations are more consistent with observed concentrations when using the local bottom-up inventory. Both emissions and simulated concentrations of PM over urban sites in Paris are much lower due to different spatial distribution of the anthropogenic emissions. The difference for nearby rural stations is small implicating that the PM urban increment is much smaller than when using the downscaled emission inventory. Urban increments for PM based on conventional down-scaled emissions may therefore be overestimated. (C) 2013 Elsevier Ltd. All rights reserved.