Dependência espacial da produtividade agrícola do feijão nos municípios cearenses nos anos de 2008 a 2016

Abstract

In 2016, the state of Ceará had the largest bean planted area with 13.0% of the country, however, participating with only 2.2% of the amount produced from this crop, resulting in the twenty-third position in national productivity. The present dissertation aims to identify the existence or not of spatial dependence of the agricultural productivity of beans from the years 2008 to 2016, considering all the municipalities of Ceará over the years. The specific objectives are to identify the existence of univariate global and local dependence of this crop as well as its dependence on temporal space, as well as to verify the existence of local patterns of dependence of the bean's agricultural productivity in relation to four selected climatic variables. To achieve these objectives, the exploratory spatial data analysis (ESDA) was used based on data from the Municipal Agricultural Survey provided by IBGE and the Cearense Foundation for Meteorology and Water Resources (Funceme). The main results point to the presence of spatial dependence in this culture in all years. However, the strength of the linear association identified was reduced in the period directly affecting the four types of local spatial patterns, better known as spatial clusters, whose main determinant factor may be below-average rainy years. Regarding the time lag, the results indicate the presence of temporal space dependence in the agricultural productivity of the beans of Ceará in all the years, whose pattern local local time was of the Low-Low (BB) type, revealing a persistent characteristic of low productivity of the beans in the great majority of municipalities of Ceará. Regarding the climatic variables, the main results point to the presence of spatial dependence of the agricultural productivity of the beans in relation to all of them in almost every year, with change of the signal and the strength of the autocorrelation differing between years above and below the average of rain. Finally, it was possible to observe several displacements of the four types of spatial clusters in the territory of the state of Ceará with different changes in their composition in terms of municipalities, but maintaining the pattern of spatial dependence in some regions, Low-Low (BB), revealing the low productivity of this crop in the State around the municipalities with the worse climatological conditions.