Atmospheric conditions important for the assessment of population exposure

Atmospheric distribution of a pollutant can be predicted using numerical weather prediction models and atmospheric dispersion models. The first provides prediction on the evaluation of the meteorological fields for specified time period and the second uses this information to determine the evolution of the dispersing cloud in time and space. There is a number of conditions and features that limit the performance of both models, as they contain a degree of parametrisation that may be a source of error. This paper discusses influential parameters and conditions.(author)