Função de produção e espacialização da precipitação efetiva: estudo de caso em fazendas de algodão e soja da Região Oeste da Bahia nas safras 2008/09 e 2009/10

• Ano de defesa: 2013
• Autor(a) principal: Bittencourt, Flávio
• Orientador(a): Mantovani, Everardo Chartuni
• Banca de defesa: Silva, José Geraldo Ferreira da , Palaretti, Luiz Fabiano , Souza, Maurício Novaes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Doutorado em Engenharia Agrícola
• Departamento: Construções rurais e ambiência; Energia na agricultura; Mecanização agrícola; Processamento de produ
• País: BR
• Palavras-chave em Português: Soja; Algodão; Precipitação pluvial
• Palavras-chave em Inglês: Soybean; Cotton; Rainfall
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://locus.ufv.br/handle/123456789/756

Under rainfed crop productivity is highly dependent on the interactions between the stages of development of culture and climate variations. The intensity, regularity and distribution of rainfall significantly interfere with this type of cultivation. It is necessary to identify the most propitious time for planting to minimize the effects of climatic variables. Data provided by IRRIGER in partnership with the group MAEDA coming from 15 farms located in the Western State of Bahia, in the municipalities of San Desiderio and Correntina during the crops 2008/09 and 2009/10 were used in this work. The actual availability of water depths 0-30 cm, 30-60 cm and 60-90 cm, planting date, effective precipitation, evapotranspiration and crop evapotranspiration from each farm were the variables adopted. As a general objective were confecionados models to represent the production functions for crops of cotton and soybeans and specific objective, we studied the spatial distribution of effective precipitation of the three wettest months of the Western Region of Bahia, December, January and February each crop. To find models that describe the productivity of the two crops linear regression was used, with the candidate variables to the regression model: actual availability of water at three depths (0-30, 30-60 and 60-90 cm), date planting, effective precipitation, evapotranspiration and crop evapotranspiration. We conducted a descriptive study, it was found collinearity of variables and, subsequently, by the method stepwise and AIC criterion met the variables that compose the models that best described the productivity for each crop and harvest. In 2008/09 for the cotton crop was not possible to adopt any model to describe productivity. However, for 2009/10 of the culture met a model with two independent variables: date of planting and effective precipitation. For soybeans in the 2008/09 planting date and crop evapotranspiration were the variables used to describe the productivity. In the crop 2009/10 for soybeans variables were the planting date and the actual availability of water in the sou l at the depth 0-30 cm. The Indices of Concordance Willmott were below 0.92. For Ra2i values were obtained less than 71 %. The rainfall data collected by rain gauges were georeferenced useful for modeling spatially effective precipitation. The models were adopted by the spherical Gaussian with better performance for the errors and the relationship observed and estimated. The month of January 2010 showed the highest degree of spatial dependence, 60%, and presented the lowest correlation between the observed and predicted values. Six-generated maps corresponding to the months December, January and February of each crop by ordinary kriging. We observed different behaviors monthly and yearly, with marked spatial and temporal variation. The interrelation between variables in each period and culture indicate that the influence of a certain parameter on productivity varies statistically in every situation.