FREITAS, Paulo Sérgio L.; GONÇALVES, Antônio Carlos A.; DALLACORT, Rivanildo.; ARALDI, Rosilaine.
Resumo:
This study was carried out at the region of Maringá, state of Paraná (latitude of
23º27 S, longitude of 51º57 W and altitude of 542 meters), where the growth and development of
the soy crop simulation model was used (Glycine Max (L) Merrill), CROPGRO-SOYBEAN, which
was calibrated to the soybean crop CD 210, cultivated in the region of Palotina, state of Paraná.
After the calibration, the simulation of the soybean productivity for five different sowing dates and
in twenty-five harvest period, from 1974 to 1999, was carried out. Through the analysis carried out
in the simulations, it was observed that the model showed a high sensibility to the variation of the
studied crops genetic coefficient. From the five dates of simulated sowing, the one that presented
the highest productivities was October 1st, but in this date we also found the lowest productivity,
October 15th and November 01st were considered the best dates to the sowing of soy crop,
presenting an average from the four crops in the twenty-five simulation of 2.450kg ha-1, and
2.303kg ha-1, respectively. The analysis of the probability of incomes demonstrates to be a great
tool in the aid the socket of decision as for the time of sowing of the culture of the soybean.
