Models which describe geophysical problems have often more than one inherent timescale. Fast timescales can be responsible for numerical stiffness of the problem. However in applications we are often not interested in the model behaviour on the fast timescale, but it is sufficient to compute a "slow solution" and apply an averaging technique to the fast frequencies.

This presentation will deal with the timescales of models describing warm clouds and ice clouds. The model behaviour of warm clouds will be investigated on different timescales by means of single scale asymptotic analysis. Furthermore the timescales of the different microphysical cloud processes will be discussed. The ice cloud model will be used to introduce a numerical averaging technique. As there are different timescales in fluid flows too, we can apply averaging to formulate the mean flows of the fluids and use these formulations for the construction of numerical algorithms.