The modelling problems associated with spatial averaging

The problem of spatial averaging in photochemical models of the stratosphere is considered. It is shown that correlation between reacting species, or between species and temperature, can lead to a bias in the model treatment of the reaction rate. Using satellite data it is demonstrated that for reactions controlling the ozone budget the resulting error is generally not large, although it may be appreciable during times of strong dynamical activity. Finally, we speculate that the errors could be more significant for species with strong gradients in the vertical, or for variability on a smaller scale than is resolved by satellite measurements. This could afford an explanation of some current modelling problems.