Modelos agrometeorológicos para previsão de atributos de qualidade em laranjas 'Valência' para o estado de São Paulo

• Ano de defesa: 2014
• Autor(a) principal: Moreto, Victor Brunini [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Laranja; Laranja - Qualidade; Previsão; Agricultura - Previsão; Meteorologia agricola; Agriculture Forecasting
• Link de acesso: http://hdl.handle.net/11449/124023

The main producer country of oranges at the world is Brazil, been the State of São Paulo responsible for almost all of this production. The 'Valência' orange is one of the most important cultivars of the State. The major influent factors on yield and quality variability of crops are climatic factors, where this annually variation can bring positive or negative effects. Forecasting is the use of actual data to estimate an event that will happen, while estimation uses actual data to esteem a current event. This study aims to understand the influence of monthly meteorological variables during the 'Valência' orange grafted on rangpur lime of orchards with more than six years old (VACR), for four main producers regions of the State of São Paulo (Bauru, Bebedouro, Limeira and Matão) and develop agrometeorological models for predict the qualitative attributes of every month of VACR during the production year. . Water deficits (DEF) are limiting factors at yield and quality of fruits. Therefore its influence on crop development was analyzed isolated; developing estimation models in function only of monthly DEF, in order to determine which phenological phases of VACR are more sensitive to DEF and with that, make accurate estimation of the qualitative attributes of VACR, at four regions of the State of São Paulo. The models were constructed in multiple linear regressions and classified according to the mean absolute perceptual error (MAPE) and the adjusted coefficient of determination (R² adjusted). All the developed models were accurate. Relative to the estimation models, for number of fruits per box, DEF were more important at the phases of the first year of the cycle and for the others attributes at the second year. For the forecasting models the main influent variables were: minimum, mean and maximum temperatures and relative evapotranspiration