Taking into account the very large area of corn crop in Brazil along side the constant demand for the improvement of crop management, there is a need for further investigation into the new methodologies for the treatment of inputs required by the production system. Maize yield studies have demonstrated significant spatial and time-related variability along the growing seasons. The simulation of the crop development may serve while discussing the variability and its effects on yield, as well as for data integration, when looking for useful information. This happens, mainly when moving from a homogeneous treatment of cultivated field toward a site-specific based management; where the difference in potential yield is due to occur, inside the same cultivated area. In accordance with the Ceres-Maize model, soil, plant and climate data, were collected in a corn growing area, located around the Angatuba (Sao Paulo), Brazil, at the Vale do Parapanema region (23º33' S; 48º18' W; 670 m). The simulation of yield related properties, for a single growing season and for a medium term climatic series of data was performed, in three separate soil units. The Ceres-Maize model allows the simulation of corn development on a daily-step basis, based upon a balance of carbon, nitrogen and water, provided that key morphogenetic traits of a given cultivar are known. Soil, plant and climate interactions implied moderate differences in the pattern of simulated corn yield variability along years, for two separate zones, inside the same cultivated area. This caused the simulated rate of nitrogen for the maximum economic return, to be moderately different. The simulation process suggested that the use of the model as a "non-point" analytical tool is conditional to the level of spatial contrasts in soil physical properties and yield potential.
Ceres-Maize; agrometeorology; soil-water; Zea mays L.
