Using the CropSyst Model in Continuous Rainfed maize (Zea mais L.) Under Alternative Management Options

BACKGROUND. Simulation models are increasingly being used for describing and predicting soil-crop variables under a wide variety of conditions in view of providing guidelines at farming level. The CropSyst model was regarded as a valuable tool for investigating continuous maize, relatively to crop production and environmental concerns. METHODS. The CropSyst model was evaluated for its ability to simulate maize biomass and nitrogen uptake, and two soil variables (i.e., water and NO3-N) in response to different types of tillage, nitrogen fertilisation and ground cover at Pisa, Central Italy. Experimental data sampled during the growing seasons 1994 and 1995 were used for this purpose. The most complex hydrological sub-models included in CropSyst were used (Penman-Monteith for evapo-transpiration, finite difference approach for soil water transport). Estimates provided by the model were compared to the state variables measured during the experiment (green area index, above ground biomass, nitrogen uptake at harvest, soil water content, and soil NO3-N in the upper 0.60 m of soil). RESULTS. CropSyst gave reasonable estimates of crop area index (average modelling efficiency, EF, equal to 0.96) and biomass (EF=0.82) and soil water content (EF=0.75). Even if the model followed the mean trend of nitrates, a few discrepancies between simulations and measurements were observed in some cases (EF=0.23). Some discrepancies were registered for plant nitrogen uptake as well (in particular in the second year of simulation), but the departures were small in many cases. CONCLUSIONS. A not fully satisfying response observed in nitrogen simulation does not appear as being extremely dramatic, considering the large variability associated to soil NO3-N, due to several casual factors influencing nitrogen variation in the soil profile. In general, the performance was appropriate and the model appeared as a valid tool for predictive purposes under different management applications. After such preliminary results, further investigation over long-term cultivation is required.