SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID

Detalhes bibliográficos
Autor(a) principal: Silva, Anderson Santos da
Data de Publicação: 2022
Outros Autores: Moura, Geber Barbosa de Albuquerque, Lopes, Pabrício Marcos Oliveira, Guimarães, Miguel Julio Machado, Bezerra, Alan Cezar, Gomes, Anthony Wellington Almeida
Tipo de documento: Artigo
Idioma: por
Título da fonte: Nativa (Sinop)
DOI: 10.31413/nativa.v10i4.13773
Texto Completo: https://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/13773
Resumo: This study aimed to build a multiple linear regression model to estimate productivity of sugarcane in the northeastern semi-arid region. Anual data of agricultural income were used (harvest 2005/2006 to 2011/2012) and monthly agro-meteorological data (2005-2012). For the model calibration period, the choice of the independent variable of the regression was analyzed by correlation of crop weather data and output data then were defined variables and constructed linear regression to estimate the yield of sugar cane -sugar. The independent variables selected for the model were irrigation more rainfall, average air temperature, the vapor saturation deficit of the air and the photoperiod. At the calibration period, the multiple linear regressions showed satisfactory results with average relative difference of less than 3%, and a standard error of estimate of 2.7264 tons of sugarcane in all crop years analyzed. Validation of the agro-meteorological model, the best performance was obtained in crop year 2004/2005 compared to the crop years of 2013/2014 and 2014/2015, respectively, a period that was renewed planting. By using a correction factor, the agro-meteorological model obtained an adjustment in crop years of 2013/2014 and 2014/2015 improving its performance. Features such as low cost, easy to implement and precision make the multiple linear regressions as an excellent tool.
id UFMT-2_f23f1c40235dd743bb14f79940bda964
oai_identifier_str oai:periodicoscientificos.ufmt.br:article/13773
network_acronym_str UFMT-2
network_name_str Nativa (Sinop)
spelling SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID PREDICCIÓN DE PRODUCTIVIDAD DE CAÑA DE AZÚCAR BASADA EN MODELO AGROMETEOROLÓGICO EN EL SEMIÁRIDO BRASILEÑOPREDIÇÃO DA PRODUTIVIDADE DE CANA-DE-AÇÚCAR COM BASE EM MODELO AGROMETEOROLÓGICO NO SEMIÁRIDO BRASILEIROPrecipitation, temperature; rainfall, water resource, weather.precipitaçãoagrometeorologiaSaccharum officinarummodelagemThis study aimed to build a multiple linear regression model to estimate productivity of sugarcane in the northeastern semi-arid region. Anual data of agricultural income were used (harvest 2005/2006 to 2011/2012) and monthly agro-meteorological data (2005-2012). For the model calibration period, the choice of the independent variable of the regression was analyzed by correlation of crop weather data and output data then were defined variables and constructed linear regression to estimate the yield of sugar cane -sugar. The independent variables selected for the model were irrigation more rainfall, average air temperature, the vapor saturation deficit of the air and the photoperiod. At the calibration period, the multiple linear regressions showed satisfactory results with average relative difference of less than 3%, and a standard error of estimate of 2.7264 tons of sugarcane in all crop years analyzed. Validation of the agro-meteorological model, the best performance was obtained in crop year 2004/2005 compared to the crop years of 2013/2014 and 2014/2015, respectively, a period that was renewed planting. By using a correction factor, the agro-meteorological model obtained an adjustment in crop years of 2013/2014 and 2014/2015 improving its performance. Features such as low cost, easy to implement and precision make the multiple linear regressions as an excellent tool.El objetivo fue construir un modelo de regresión lineal múltiple para estimar la productividad de la caña de azúcar en la región semiárida nororiental. Se utilizaron datos de rendimiento agrícola anual (cosechas 2005/2006 a 2011/2012) y datos agrometeorológicos mensuales (2005 a 2012). Para calibrar el modelo, se analizó la elección de variables independientes a través de la correlación existente de datos agrometeorológicos y rendimiento, luego se construyó una regresión lineal múltiple para estimar la productividad de la caña de azúcar. Las variables independientes seleccionadas para el modelo fueron riego más lluvia, temperatura media del aire, déficit de saturación de vapor de aire y fotoperiodo. Durante el período de calibración, las regresiones lineales múltiples mostraron resultados satisfactorios con una diferencia relativa promedio inferior al 3% y un error estándar de estimación de 2,7264 toneladas de caña de azúcar en todos los años de cosecha analizados. En la validación del modelo agrometeorológico, el mejor desempeño se obtuvo en la campaña 2004/2005 en comparación con las campañas 2013/2014 y 2014/2015, respectivamente, período en el que se renueva la siembra. Mediante el uso de un factor de corrección, el modelo agrometeorológico obtuvo un ajuste en las campañas 2013/2014 y 2014/2015, mejorando su desempeño. Características como el bajo costo, la facilidad de ejecución y la precisión hacen de las regresiones lineales múltiples una excelente herramienta.A predição da produtividade de cultivos de cana-de-açúcar é fundamental para o planejamento logístico dos produtores. No entanto, muitas propriedades recorrem a estimativa visual de produtividades em áreas canavieiras, o que, muitas vezes, geram dados tendencioso que não refletem a real produtividade dos cultivos. Objetivou-se construir um modelo de regressão linear múltipla para a estimativa de produtividade da cana-de-açúcar na região do semiárido nordestino. Foram utilizados dados anuais de rendimento agrícola (safras 2005/2006 a 2011/2012) e agrometeorológicos mensais (2005 a 2012). Para a calibração do modelo, a escolha das variáveis independentes foi analisada mediante a correlação existente dos dados agrometeorológicos e de rendimento, logo em seguida foi construída a regressão linear múltipla para estimar a produtividade da cana-de-açúcar. As variáveis independentes selecionadas para o modelo foram a irrigação mais precipitação pluvial, a temperatura média do ar, o déficit de saturação de vapor do ar e o fotoperíodo. No período da calibração, as regressões lineares múltiplas apresentaram resultados satisfatórios com diferença relativa média inferior a 3% e um erro padrão de estimativa de 2,7264 toneladas de cana em todos os anos-safras analisados. Na validação do modelo agrometeorológico, o melhor desempenho foi obtido no ano-safra de 2004/2005 quando comparado com os anos-safras de 2013/2014 e 2014/2015, respectivamente, período que houve a renovação de plantio. Mediante o uso de um fator de correção, o modelo agrometeorológico obteve um ajuste nos anos-safras de 2013/2014 e 2014/2015 melhorando seu desempenho. Características como baixo custo, facilidade de executar e precisão tornam as regressões lineares múltiplas como excelentes ferramentas. Palavras-chave: precipitação; Saccharum officinarum; modelagem; agrometeorologia.   Sugarcane yield prediction based on agrometeorological model in the Brazilian semi-arid   ABSTRACT: The prediction of the productivity of sugarcane crops is essential for the logistical planning of producers. However, many properties resort to visual estimates of yields in sugarcane areas, which often generate biased data that do not reflect the real productivity of crops. This study aimed to build a multiple linear regression model to estimate productivity of sugarcane in the northeastern semi-arid region. Anual data of agricultural income were used (harvest 2005/2006 to 2011/2012) and monthly agro-meteorological data (2005-2012). For the model calibration period, the choice of the independent variable of the regression was analyzed by correlation of crop weather data and output data then were defined variables and constructed linear regression to estimate the yield of sugar cane -sugar. The independent variables selected for the model were irrigation more rainfall, average air temperature, the vapor saturation deficit of the air and the photoperiod. At the calibration period, the multiple linear regressions showed satisfactory results with average relative difference of less than 3%, and a standard error of estimate of 2.7264 tons of sugarcane in all crop years analyzed. Validation of the agro-meteorological model, the best performance was obtained in crop year 2004/2005 compared to the crop years of 2013/2014 and 2014/2015, respectively, a period that was renewed planting. By using a correction factor, the agro-meteorological model obtained an adjustment in crop years of 2013/2014 and 2014/2015 improving its performance. Features such as low cost, easy to implement and precision make the multiple linear regressions as an excellent tool. Keywords: precipitation; Saccharum officinarum; modeling; agrometeorology.Universidade Federal de Mato Grosso2022-11-22info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/1377310.31413/nativa.v10i4.13773Nativa; v. 10 n. 4 (2022); 515-524Nativa; Vol. 10 Núm. 4 (2022); 515-524Nativa; Vol. 10 No. 4 (2022); 515-5242318-7670reponame:Nativa (Sinop)instname:Universidade Federal de Mato Grosso (UFMT)instacron:UFMTporhttps://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/13773/11635Copyright (c) 2022 Nativahttps://creativecommons.org/licenses/by-nc/4.0info:eu-repo/semantics/openAccessSilva, Anderson Santos daMoura, Geber Barbosa de AlbuquerqueLopes, Pabrício Marcos OliveiraGuimarães, Miguel Julio MachadoBezerra, Alan CezarGomes, Anthony Wellington Almeida 2022-11-02T00:23:39Zoai:periodicoscientificos.ufmt.br:article/13773Revistahttps://periodicoscientificos.ufmt.br/ojs/index.php/nativaPUBhttps://periodicoscientificos.ufmt.br/ojs/index.php/nativa/oai||rrmelo2@yahoo.com.br2318-76702318-7670opendoar:2022-11-02T00:23:39Nativa (Sinop) - Universidade Federal de Mato Grosso (UFMT)false
dc.title.none.fl_str_mv SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
PREDICCIÓN DE PRODUCTIVIDAD DE CAÑA DE AZÚCAR BASADA EN MODELO AGROMETEOROLÓGICO EN EL SEMIÁRIDO BRASILEÑO
PREDIÇÃO DA PRODUTIVIDADE DE CANA-DE-AÇÚCAR COM BASE EM MODELO AGROMETEOROLÓGICO NO SEMIÁRIDO BRASILEIRO
title SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
spellingShingle SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
Silva, Anderson Santos da
Precipitation, temperature; rainfall, water resource, weather.
precipitação
agrometeorologia
Saccharum officinarum
modelagem
Silva, Anderson Santos da
Precipitation, temperature; rainfall, water resource, weather.
precipitação
agrometeorologia
Saccharum officinarum
modelagem
title_short SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
title_full SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
title_fullStr SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
title_full_unstemmed SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
title_sort SUGARCANE YIELD PREDICTION BASED ON AGROMETEOROLOGICAL MODEL IN THE BRAZILIAN SEMI-ARID
author Silva, Anderson Santos da
author_facet Silva, Anderson Santos da
Silva, Anderson Santos da
Moura, Geber Barbosa de Albuquerque
Lopes, Pabrício Marcos Oliveira
Guimarães, Miguel Julio Machado
Bezerra, Alan Cezar
Gomes, Anthony Wellington Almeida
Moura, Geber Barbosa de Albuquerque
Lopes, Pabrício Marcos Oliveira
Guimarães, Miguel Julio Machado
Bezerra, Alan Cezar
Gomes, Anthony Wellington Almeida
author_role author
author2 Moura, Geber Barbosa de Albuquerque
Lopes, Pabrício Marcos Oliveira
Guimarães, Miguel Julio Machado
Bezerra, Alan Cezar
Gomes, Anthony Wellington Almeida
author2_role author
author
author
author
author
dc.contributor.author.fl_str_mv Silva, Anderson Santos da
Moura, Geber Barbosa de Albuquerque
Lopes, Pabrício Marcos Oliveira
Guimarães, Miguel Julio Machado
Bezerra, Alan Cezar
Gomes, Anthony Wellington Almeida
dc.subject.por.fl_str_mv Precipitation, temperature; rainfall, water resource, weather.
precipitação
agrometeorologia
Saccharum officinarum
modelagem
topic Precipitation, temperature; rainfall, water resource, weather.
precipitação
agrometeorologia
Saccharum officinarum
modelagem
description This study aimed to build a multiple linear regression model to estimate productivity of sugarcane in the northeastern semi-arid region. Anual data of agricultural income were used (harvest 2005/2006 to 2011/2012) and monthly agro-meteorological data (2005-2012). For the model calibration period, the choice of the independent variable of the regression was analyzed by correlation of crop weather data and output data then were defined variables and constructed linear regression to estimate the yield of sugar cane -sugar. The independent variables selected for the model were irrigation more rainfall, average air temperature, the vapor saturation deficit of the air and the photoperiod. At the calibration period, the multiple linear regressions showed satisfactory results with average relative difference of less than 3%, and a standard error of estimate of 2.7264 tons of sugarcane in all crop years analyzed. Validation of the agro-meteorological model, the best performance was obtained in crop year 2004/2005 compared to the crop years of 2013/2014 and 2014/2015, respectively, a period that was renewed planting. By using a correction factor, the agro-meteorological model obtained an adjustment in crop years of 2013/2014 and 2014/2015 improving its performance. Features such as low cost, easy to implement and precision make the multiple linear regressions as an excellent tool.
publishDate 2022
dc.date.none.fl_str_mv 2022-11-22
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/13773
10.31413/nativa.v10i4.13773
url https://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/13773
identifier_str_mv 10.31413/nativa.v10i4.13773
dc.language.iso.fl_str_mv por
language por
dc.relation.none.fl_str_mv https://periodicoscientificos.ufmt.br/ojs/index.php/nativa/article/view/13773/11635
dc.rights.driver.fl_str_mv Copyright (c) 2022 Nativa
https://creativecommons.org/licenses/by-nc/4.0
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2022 Nativa
https://creativecommons.org/licenses/by-nc/4.0
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade Federal de Mato Grosso
publisher.none.fl_str_mv Universidade Federal de Mato Grosso
dc.source.none.fl_str_mv Nativa; v. 10 n. 4 (2022); 515-524
Nativa; Vol. 10 Núm. 4 (2022); 515-524
Nativa; Vol. 10 No. 4 (2022); 515-524
2318-7670
reponame:Nativa (Sinop)
instname:Universidade Federal de Mato Grosso (UFMT)
instacron:UFMT
instname_str Universidade Federal de Mato Grosso (UFMT)
instacron_str UFMT
institution UFMT
reponame_str Nativa (Sinop)
collection Nativa (Sinop)
repository.name.fl_str_mv Nativa (Sinop) - Universidade Federal de Mato Grosso (UFMT)
repository.mail.fl_str_mv ||rrmelo2@yahoo.com.br
_version_ 1822182913919156224
dc.identifier.doi.none.fl_str_mv 10.31413/nativa.v10i4.13773

Registros relacionados