Desempenho do modelo CROPGRO-Dry bean em estimar a data de semeadura e a produtividade do feijoeiro

• Autor(a) principal: Oliveira, Evandro Chaves de
• Data de Publicação: 2007
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/5286
• topic: Feijão; Semeadura; Produtividade; Common bean; Sowing; Productivity; CNPQ::CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS::METEOROLOGIA::MICROMETEOROLOGIA

Crop simulation models are tools that generate sceneries, considering the several combinations of elements that influence on crop productivity. Simulation models have been used for optimization of managements practices as well as to estimate crop yield. The CROPGRO model is widely applied in crop productivity estimates. The objectives of the present study are: 1) to evaluate the performance of the model CROPGRO-Dry bean for estimating the growing stages and the productivity of the common bean (Phaseolus vulgaris L.), in Viçosa, MG; 2) to apply the model for the determination of the best sowing dates of bean, for rainfed conditions; 3) to test the model in the prediction of the productivity of the common bean. The adjustment of the genetic coefficients of the model was done to Pérola, Ouro Negro and Ouro Vermelho bean cultivars, in Viçosa-MG. The coefficients are determined based on experimental data. Those data were obtained from two experiments in the year of 2003, for two different conditions: irrigation and rainfed. An additional experiment was conducted in the year of 2004, with irrigation. After calibration, the model simulated the bean productivity for 31 years between 1975 and 2006. The simulations applied climatological daily of maximum and minimum temperatures, total precipitation and global solar radiation; physical hydric characteristics of the soil and crop management practice. According to the results, the model is highly sensitive to genetic coefficient variations between the three distinct cultivars during the growth stages and grain productivity. CROPGRO-Dry bean model simulated with accuracy the bean phenology based on the soil and climate conditions of the Viçosa-MG. It was simulate satisfactorily the grain productivity, since values of RMSE were less than 5 % for Pérola and Ouro Negro and, 12, 63 %, Ouro Vermelho cultivars. In general, for the two sceneries, potential and real, as more delayed the planting date is, lower is the productivity of the bean for the dry season for the three cultivars. The best sowing dates were found to be between October 1 to November 20. CROPGRO-Dry bean model is, therefore, a useful tool for predicting the bean yield, thirty days in advance harvest.