Validation of Goudriaan's model: a case study for maize.

The crop microclimate model of Goudriaan was tested, using data collected in a maize field in the Netherlands during one day in summer 1986. Except for the soil heat flux, latent and sensible heat fluxes were simulated reasonably well. Goudriaan's model overestimated the latent and sensible heat fluxes above the crop canopy by 9 and 10%, respectively, in comparison with the eddy correlation technique. The simulated fluxes could be improved by a better incorporation of the effect of dewfall in the model. Root mean square errors of the simulated soil temperatures at 1, 5, 10 and 20 cm depth were between 0.30 and 0.68 K. Sensitivity analysis of the model showed that the heat fluxes were more sensitive to soil than to crop parameters, for the same magnitude of variations. Among soil parameters, the soil surface resistance to evaporation had a large influence on the latent heat and the sensible heat fluxes at the soil surface; it also influenced the soil heat flux. Soil water stress had a large influence on the daily net CO2 assimilation as well as the latent and the sensible heat fluxes above the crop canopy.