MODELOS AGROMETEOROLÓGICOS PARA PREVISÃO DA PRODUÇÃO DE MILHO EM MATO GROSSO DO SUL

Autores

  • MARCO AURÉLIO ARGENTA MOCINHO JUNIOR
  • WILLIAN PEREIRA CENTURION
  • ARTHUR FERREIRA SOUZA PRADO
  • GUILHERME BOTEGA TORSONI
  • LUCAS EDUARDO DE OLIVEIRA APARECIDO
  • CICERO TEIXEIRA SILVA COSTA

DOI:

https://doi.org/10.15809/irriga.2019v1n1p38-47

Resumo

MODELOS AGROMETEOROLÓGICOS PARA PREVISÃO DA PRODUÇÃO DE MILHO EM MATO GROSSO DO SUL

 

 

MARCO AURÉLIO ARGENTA MOCINHO JUNIOR1; WILLIAN PEREIRA CENTURION1; ARTHUR FERREIRA SOUZA PRADO1; GUILHERME BOTEGA TORSONI2; LUCAS EDUARDO DE OLIVEIRA APARECIDO2 E CICERO TEIXEIRA SILVA COSTA2

 

1Estudante do curso de Engenharia Agronômica do IFMS campus Naviraí. Laboratório de Engenharia Agrícola, Rua Hilda, 203, Naviraí - MS. CEP. 79950-000, E-mail: marcoaurelio18@live.com

2Docentes do IFMS campus Naviraí. Rua Hilda,203, Naviraí-MS. CEP. 79950-000, E-mail: cicero.costa@ifms.edu.br

 

 

1 RESUMO

 

O milho representa um dos principais cereais cultivado e consumido no mundo, em virtude do seu alto potencial produtivo, composição química e valor nutritivo. No entanto, a sua produção é altamente dependente do clima. Objetivou-se estimar a produção do milho por meio da calibração de modelos estatísticos para o Estado de Mato Grosso do Sul - MS. As cidades estudadas foram Chapadão do Sul, Costa Rica, Ponta Porã e Sidrolândia. As variáveis climáticas utilizadas foram temperatura do ar, a precipitação pluvial, evapotranspiração potencial, déficit e o excesso hídrico no período de 2003 - 2017 entre fevereiro e maio. Os modelos foram calibrados e comparados pelos métodos KNN e RANDOM. A acurácia e a precisão dos modelos foram analisadas pelo erro percentual médio e pelo coeficiente de determinação ajustado, respectivamente. As variáveis que mais influenciaram na produção do milho foram o déficit hídrico e a temperatura do ar. É possível estimar a produção do milho com regressões lineares múltiplas utilizando variáveis climáticas. Chapadão do Sul e Costa Rica apresentam altos índices de déficit hídrico, enquanto Ponta Porã e Sidrolândia baixos déficits. O modelo mais acurado para estimar a produção do milho nas cidades foi o método RANDOM.

 

Keywords: Clima; Produtividade; Modelagem.

 

 

MOCINHO, M. A. A.; CENTURION, W. P; PRADO, A. F. S; TORSONI, G. B; APARECIDO, L. E. O; COSTA, C. T. S

AGROMETEOROLOGICAL MODELS FOR FORECASTING MAIZE PRODUCTION IN MATO GROSSO DO SUL

 

 

2 ABSTRACT

 

Corn represents one of the main cereals cultivated and consumed in the world, due to its high productive potential, chemical composition and nutritional value. However, its production is highly climate dependent. The objective of this study was to estimate maize yield by calibrating statistical models for the state of Mato Grosso do Sul - MS. The cities studied were Chapadão do Sul, Costa Rica, Ponta Porã and Sidrolândia. The climatic variables used were air temperature, rainfall, potential evapotranspiration, deficit and excess water from 2003 to 2017 between February and May. The models were calibrated and compared by the KNN and RANDOM methods. The accuracy and precision of the models were analyzed by the mean percentage error and the adjusted determination coefficient, respectively. The variables that most influenced corn production were water deficit and air temperature. It is possible to estimate corn yield with multiple linear regressions using climate variables. Chapadão do Sul and Costa Rica have high levels of water deficit, while Ponta Porã and Sidrolândia have low deficits. The most accurate model for estimating maize yield in cities was the RANDOM method.

 

Keywords: Climate; Yield; Modeling.

 

Downloads

Publicado

2019-10-30

Como Citar

MOCINHO JUNIOR, M. A. A. .; CENTURION, W. P. .; PRADO, A. F. S. .; TORSONI, G. B. .; APARECIDO, L. E. D. O. .; COSTA, C. T. S. . MODELOS AGROMETEOROLÓGICOS PARA PREVISÃO DA PRODUÇÃO DE MILHO EM MATO GROSSO DO SUL. IRRIGA, [S. l.], v. 1, n. 1, p. 38–47, 2019. DOI: 10.15809/irriga.2019v1n1p38-47. Disponível em: https://irriga.fca.unesp.br/index.php/irriga/article/view/3888. Acesso em: 26 abr. 2024.