Módulos computacionais de análise geoestatística e retificação de zonas de manejo

• Ano de defesa: 2017
• Autor(a) principal: Betzek, Nelson Miguel
• Orientador(a): Souza, Eduardo Godoy de
• Banca de defesa: Opazo, Miguel Angel Uribe , Pinheiro Neto, Raimundo , Gonçalves, Antonio Carlos Andrade , Maggi, Marcio Furlan
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual do Oeste do Paraná; Cascavel
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Agrícola
• Departamento: Centro de Ciências Exatas e Tecnológicas
• País: Brasil
• Palavras-chave em Português: Agricultura de precisão; Índice de suavização; Krigagem ordinária; Sistemas computacionais
• Palavras-chave em Inglês: Computer systems; Precision agriculture; Ordinary kriging; Smoothing
• Área do conhecimento CNPq: CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://tede.unioeste.br/handle/tede/3082

The use of computer systems for the management of agricultural procedures has been an emerging need. However, applications used for this purpose generally present restrictions on functionality, operating licenses, operating system and others. Thus, the Software for the Definition of Management Units (SDUM) was developed. This thesis aimed at implementing computational routines that will be integrated into SDUM, which are able to identify automatically the best parameters for the ordinary kriging (KRI) interpolation methods and inverse distance weighting (IDW). These routines were applied to the sample data of selected attributes to define MZs in two agricultural areas. For each dataset 300 different adjustments were tested for the semivariogram. The best parameters were used to measure data by KRI as well as twelve different values for IDW exponent. Area A was considered homogeneous because it did not present statistic different mean between classes. While for area B, the best result was subdivided into two distinct classes. MZs usually present isolated cells or patches, in such a way that crop management becomes difficult. In this context, another goal was to implement routines able of adjusting MZs, to smooth and to improve continuity. Three sampled areas were subdivided into two, three, four and five classes, and then adjusted for five times. Isolated pixels and almost all patches were eliminated, especially the methods with 5 × 5 mask. So, it can be concluded that the computational routines implemented were efficient and able to identify the best adjustment for the semivariogram as well as the best exponent for IDW, also smoothing and improving MZs continuity.