Parameterization of a crop growth and development simulation model at sub-model components level. An example for winter wheat (Triticum aestivum L.)

Dynamic simulation models are frequently used for assessing agronomic and environmental effects of different management practices, under various pedo-climatic conditions. CropSyst is a suitable cropping systems simulation model for such applications. However, available CropSyst crop parameters for winter wheat, one of the most important cereals in the world, are limited. In this work we show that it is possible to parameterize separate sub-model components by using existing experimental data and literature. The experiments, carried out in northern Italy between 1986 and 2001, quantified the dynamics of aboveground biomass (AGB), plant nitrogen (N) concentration (PNC) and N uptake (UPTK) by means of periodical measurements. The relative root mean square error (calculated by dividing the root mean square error by the average of observations) obtained after model calibration and validation on an independent data set was, respectively, in the range 9-30% and 17-32% for AGB, 10% and 6-40% for PNC, 8-28% and 9-24% for UPTK. AGB was frequently underestimated. Despite the limited accuracy of simulations, we argue that calibrated crop parameters are adequate for scenario analysis as most differences between years and fertilization levels were reproduced by the model and final AGB and cumulative UPTK were also correctly simulated.