Tese
Rendimento de grãos e de óleo do crambe em um latossolo: análise espacial e modelos de correlação
Fecha
2014-11-28Registro en:
MARINS, Araceli Ciotti de. Grains yield and oil content of crambe in an oxisol: Spatial analisys and corelation models. 2014. 161 f. Tese (Doutorado em Agronomia) - Universidade Federal de Santa Maria, Santa Maria, 2014.
Autor
Marins, Araceli Ciotti de
Institución
Resumen
The growing environmental awareness regarding production and use of renewable fuels has led many of countries to create policies to benefit producers of renewable fuels. Thus, the search for raw materials for production of biofuels, which do not conflict with the global food production and exhibit similar performance to fossil fuels, has generated interest towards crambe, a crop of high oil content, inappropriate for animal consumption and which can be applied in crop rotation without the need for exchange farm machinery. However, studies evaluating the influence of spatial variability of soil chemical and physical properties on grain yield and oil content of crambe are scarce. Thus, this study aimed to evaluate the spatial correlation between physical and chemical properties of an Oxisol under compaction states with grain yield and oil content of crambe, through a cross-correlation estimator based on moving windows and assess adequate sampling density for application of geostatistics. For this, we used geostatistical techniques such as kriging and simulation data and using software R. We concluded that increased bulk density and soil resistance to penetration change the source:sinc relationship of crambe, reflecting in lower yield but with grain production of higher quality; chemical attributes that have higher direct spatial relationship to grain yield of crambe are phosphorus, calcium, magnesium and organic matter; sampling grids constructed with points spaced at large distances are not effective in detecting the spatial variability of chemical attributes and grain yield and oil content of crambe; and that the cross semivariogram based on moving windows detects the structure of spatial correlation between the physical and chemical soil properties, independent of its variability or dispersion, showing superior performance when the data have outliers and do not have normal distribution.