MODELING CROP GROWTH AND BIOMASS PARTITIONING TO SHOOTS AND ROOTS IN RELATION TO NITROGEN AND WATER AVAILABILITY, USING A MAXIMIZATION PRINCIPLE - II - SIMULATION OF CROP NITROGEN-BALANCE

This study analysed of the ability of a crop model to simulate crop ni trogen (N) balance. The model was originally developed to serve as a f oundation to develop a decision-making tool to analyse the impact of w ater management and nitrogen fertilization on crop yield. The model in cluded a dynamic parameter for allocation of dry matter between root a nd shoot allowing root to shoot ratio to vary according to differing e nvironmental conditions. The new allocation parameter was introduced i n order to make the model more applicable under water and nitrogen lim ited growing conditions. Two wheat (Triticum aestivum L.) data sets we re used to test the model simulations. Generally, the model simulation s agreed well with the recorded data on crop N uptake. The relationshi p between the actual and simulated amount of N taken up by the crop wa s close in the calibration treatments of a greenhouse experiment. The coefficient of determination (r(2)) of the regression line (simulated value = independent variable, measured value = dependent variable) was 0.90. The r(2) was 0.83 for the validation data. In the field experim ents, the r(2) values were 0.91 for the calibration data and 0.82 for the validation data. In field data, the model underestimated in some c ases the crop N uptake during the period when actual shoot dry weight increased exponentially in spring. Therefore, methods used in computat ion of nitrogen uptake have to be analysed further. Plant organ N cont ent was simulated satisfactorily for both greenhouse and held data. Ho wever, the range over which the simulated values varied was larger tha n in the actual data. The results from the study indicate that our mod el is capable of simulating the crop N balance and we suggest that the model could be used when developing an N application decision tool fo r field crops. However, the availability of N and soil water were prov ided as inputs in the present study. Thus, the model should be integra ted with models simulating below ground processes in the future. Moreo ver, the model should be further validated with actual field data. (C) 1998 Elsevier Science B.V.