CORREÇÃO DE DADOS AGROMETEOROLÓGICOS UTILIZANDO MÉTODOS ESTATÍSTICOS

• Ano de defesa: 2012
• Autor(a) principal: Baba, Ricardo Kazuo
• Orientador(a): Vaz, Maria Salete Marcon Gomes
• Banca de defesa: Sunye, Marcos Sfair , Ribeiro, Selma Regina Aranha
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: UNIVERSIDADE ESTADUAL DE PONTA GROSSA
• Programa de Pós-Graduação: Programa de Pós Graduação Computação Aplicada
• Departamento: Computação para Tecnologias em Agricultura
• País: BR
• Palavras-chave em Português: preenchimento de falhas; dados meteorológicos; estatística; geoestatística; controle de qualidade de dados; banco de dados espacial
• Palavras-chave em Inglês: gap filling; meteorological data; statistics; geostatistics; data quality control; spatial database
• Área do conhecimento CNPq: CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://tede2.uepg.br/jspui/handle/prefix/149

Climatic data are more and more important to predict climate phenomena or to evaluate historical data that serve as support for decision making especially for agriculture. Ensuring the quality of these data is crucial. These data are collected by the meteorological stations, during this process some data gaps and data inconsistent may be generated. Identify suspicious or inconsistent data is very important to ensure data quality. This paper presents an approach that uses statistical and geostatistical techniques to identify incorrect and suspicious data and estimate new values to fill gaps and errors. In this research, a spatial database was used to implement these techniques (statistical and geostatistical) and to test and evaluate the weather data. To evaluate these techniques we used data from stations located in Paraná State to evaluate the temperature variable. To check the results of the estimated data, we used the mean absolute error (MAE) and the root mean square error (RMSE). As a result, the uses of these techniques have proved to be suitable to identify basic errors and historical errors. The temporal validation showed a poor performance by overestimating the amount of incorrect data. Regarding the estimation techniques applied Kriging, Inverse of Distance Weighted and Linear Regression, all showed similar performance in the error analysis.