An optimised method to couple meteorological and photochemical models

Accuracy of photochemical air quality model results depends not only on the accuracy of the meteorological fields supplied to the model, but also the frequency that those fields are updated. Tests in two and three dimension, with and without chemistry, were run to identify errors that can arise due to the frequency that models are linked (or coupled) to time varying meteorological fields. In applications using photochemical models uncoupled from meteorological models, the typical frequency that fields are updated is 1 h. This may lead to some error. When run in a coupled fashion, advecting trace contaminants using the same time step as the meteorological model requires excessive computational time, and does not improve results significantly beyond that found when using the CFL limit to determine the chemical species advection step.