A simulation in a vertical column was carried out to test the proper behavior of the soil and the deposition modules. No horizontal transport effect was considered. The results were compared to the eddy covariance measurements of a meadow on the Swiss Plateau in May 1993 (Eugster; 1994). Initial temperatures and wind speed were set to measured values. Figure 6 shows the comparisons between measured and modeled values. During the day the comparisons show a good agreement of all parameters, except deposition. The deposition velocity for NO 2 is slightly overestimated in the morning and in the evening while peak daytime values are underestimated. The largest deviations are found for air temperature and surface temperature of foliage in the stably stratified surface layer during night time, where the parameterization of Businger et al. (1971) used in the model predicts a very stable stratification with vanishing turbulent fluxes. Under such circumstances the soil energy balance is decoupled from the atmosphere, an effect that could not be observed in the measured data. Several authors have mentioned that Businger et al.'s (1971) parameterization in combination with a numerical model that uses a finite resolution in the vertical direction, leads to an underestimation of vertical fluxes during night-time (e.g. Louis; 1979). This problem will be addressed in future versions of Metphomod .
