Transferring the heterogeneity of surface emissions to variability in pollutant concentrations over urban areas through a chemistry-transport model

Horizontal resolution of grid-based chemistry-transport models is limited to a few square kilometers which has been proved insufficient for assessing human exposure and health impact. We propose a general methodology, applicable on any kind of grid-based air-quality model, that combines subgrid scale information on emission and land-use data in order to disaggregate the grid-averaged emission flux into a set of source-specific components (subgrid-environments). Different subgrid concentrations are calculated inside each one of these environments providing a direct estimate of pollutant variability along with the 'standard' grid-averaged model output. The method was first validated over a controlled emissions case by comparing concentrations modeled in the subgrid-environments with concentrations modeled directly at higher model resolution and next over a real case-study, where subgrid concentrations were compared with monitor data from sites representing different types of urban environments (i.e. roads and residential blocks). It was shown that the method is capable to yield accurate estimates of small scale pollutant variability.