Modelo de regressão espacial para estimativa da produtividade da soja associada a variáveis agrometeorológicas na região oeste do estado do Paraná

Abstract

RESUMO:Este trabalho apresenta o Modelo de Regressão Espacial Autorregressivo Misto (SAR) e Modelo do Erro Espacial (CAR) no intuito de investigar a associação entre a produtividade da soja e as variáveis agrometeorológicas relacionadas à precipitação pluvial, temperatura média e radiação solar global. O estudo foi realizado com os dados das safras dos anos agrícolas de 2005/2006 a 2007/2008, da região oeste do estado do Paraná. Como os dados agrometeorológicos estão disponíveis apenas para oito municípios da região em estudo, as estimativas foram obtidas por meio do uso de Polígonos de Thiessen. A estimativa de parâmetros dos modelos ajustados foi obtida utilizando o método de Máxima Verossimilhança. A avaliação do desempenho dos modelos foi realizada com base no coeficiente de determinação (R 2 ), no máximo valor do logaritmo da função verossimilhança e no critério de informação bayesiano de Schwarz (BIC). Este estudo também permitiu verificar a correlação e autocorrelação espacial entre a produtividade da soja e os elementos agrometeorológicos, por meio da análise espacial de área, usando de técnicas como o índice I de Moran Global e Local uni e bivariado, e os testes de significância. O estudo pôde demonstrar que, por meio dos indicadores de desempenho utilizados, os modelos SAR e CAR ofereceram melhores resultados em relação ao modelo de regressão múltipla clássica. As the agrometeorological data are available only for eight cities of the region in study, the estimates were obtained through the use of Thiessen polygons. The estimation of the parameters of the adjusted models was obtained using the method of maximum likelihood. The evaluation of the performance of models was held based on the coefficient of determination (R²), maximum value of the logarithm of the likelihood function and Bayesian Information Criterion of Schwarz (BIC). This study also allowed to verify the correlation and the spatial autocorrelation between soybean yield and the agrometeorological factors by analyzing spatial area, by uses of Global and Local uni and bivariate and significance tests. The study demonstrated that by means of performance indicators used, the SAR and CAR models offered better results than the classical multiple regression model. PALAVRAS-CHAVE: autocorrelação espacial, estatística espacial de área, modelos espaciais SAR e CAR. SPATIAL REGRESSION MODEL FOR SOYBEAN CROP IN THE WESTERN

Analysis of Spatial Autocorrelation of Grain Production and Agricultural Storage in Paraná

“…The largest spatial autocorrelations were found for the 2009/2010, 2010/2011 and 2013/2014 harvest years, with support for the 2010/2011 harvest year, which presented the highest values of spatial autocorrelation, indicating that there was a greater similarity of TPHG among the municipalities in this harvest, when compared to other years studied (Table 1). This result corroborates with Araújo et al (2014) who found values of the global Moran (I) indexes that ranged from 0.2203 to 0.8359 (between the harvest years from 2005/2006 to 2007/2008) in the soybean productivity study in the West region of Paraná.…”

Spatial and Multivariate Analysis of Soybean Yield in the State of Paraná-Brazil

“…However, when the 2010/2011 indices were analyzed, the correlations between the ISSN 2166-0379 2020 municipalities were more heterogeneous in relation to the 2012/2013 crop year. It is identified, therefore, that from the 2 nd ten-day period of January, the crop enters the phase of senescence and harvest, which corroborate Araújo et al (2014).…”

“…Due to the complex interrelationships that soybean yield has with other factors of this agricultural activity, it is difficult to define standards for their respective production areas (Araújo et al, 2013). Studies in the literature demonstrate the importance of the investigation regarding the existence of a relationship or an effect of agrometeorological characteristics with soybean yield, aiming to establish a better management of the production of this agricultural commodity (Araújo et al, 2013;Araújo et al, 2014;Bilbas et al, 2017).…”

“…In addition to yield information, vegetation indices have been employed as indicators to discriminate areas with soybean cultivation (Grzegozewski et al, 2016), as well as areas with other types of land use (Araújo et al, 2014), and to analyze the conditions of soybean crop growth on a regional scale.…”

Estimation of annual yield and quality of Valncia orange related to monthly water deficiencies

“…Bozorg et al (2011) developed a model for wheat crop at Irã region. Araújo et al (2014) made a modeling in order to estimate soybean yield associated to agrometeorological variables at Paraná. Guimarães et al (2013), elaborated models in multiple linear regressions to predict the banana "Prata" harvest time, at Guanambi, Bahia (BA) in function of yield characteristics, for example number of fruits.…”