Análise da evolução espaço-temporal das lavouras no Estado de Minas Gerais

• Ano de defesa: 2010
• Autor(a) principal: Delgado, Rafael Coll
• Orientador(a): Sediyama, Gilberto Chohaku
• Banca de defesa: Andrade, Ricardo Guimarães , Hamakawa, Paulo José , Justino, Flávio Barbosa
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Doutorado em Meteorologia Agrícola
• Departamento: Agrometeorologia; Climatologia; Micrometeorologia
• País: BR
• Palavras-chave em Português: Climatologia; Sensoriamento remoto
• Palavras-chave em Inglês: Climatologia; Sensoriamento remoto
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: http://locus.ufv.br/handle/123456789/1501

This work aimed to carry out the spectral classification of planted areas (coffee, sugarcane, corn, soybean and dry bean crops) by means of the decision tree classifier (DT), to analyze influence of the climate, soil and terrain relief on Coffea arabica plantations and change in the agricultural composition by means of the crops substitution index (SI). For the spectral classification of the planted areas, the algorithm (DT) and products of the TM sensor, aboard of the Landsat-5 satellite, were used. For an updated climatic zoning of Coffea arabica plantations, 1941-2009 climatic data sets were used. To analyze agricultural composition change, 1990-2008 census data, provided by the IBGE, were also used. Spectral classification by means of DT algorithm showed that the remote sensing image obtained by the Landsat-5 satellite, in four different years and specific times, allowed to map the areas occupied by the crops (coffee, sugarcane, corn, soybean, and dry bean) in the state of Minas Gerais. The planted areas, estimated by DT algorithm, presented a good correlation with the values observed by the IBGE, but there was a tendency to overestimate. When comparing the planted areas estimated via DT algorithm with those observed IBGE data coherence was verified in the results found and also supported by the literature. The Coffea arabica climatic zoning update was found to be very efficient in delimiting the climatically homogenous regions, for the potential coffee yield in the state of Minas Gerais. The methodologies to obtain an updated data set and processing of climatic zoning for Coffea arabica productivity and its suitability were shown to be technically acceptable compared to the zonings previously carried out in the state of Minas Gerais. The most outstanding crops in terms of area expansion were found to be sugarcane, coffee, soybean and dry bean, indicating that the state of Minas Gerais presented an important economic and relatively homogenous dynamism, not dependent exclusively on the performance of a single product. Thus, dissemination of the methodologies and technologies developed is recommended in order to monitor planted areas in the state of Minas Gerais.