Several advances have been made in the field of diffuse spectroradiometry in the last decades. In agriculture, the search for quantification methods of physical and chemical properties of the production environment, based on soil reflectance, has been constantly researched. The aim of this study was to evaluate the possibility of estimating the silt, sand, clay, sum of bases, total iron and organic matter content in soil samples based on the reflected energy. The samples were collected in a 2500 ha area, located in the northwestern region of Paraná, formed by Oxisol, Alfisol, Ultisol, Inceptsol, and Quartzipsamment soils, from the layers 0-0.2 and 0.2-0.8 m. The points to be sampled were chosen based on a 500 x 500 m grid, which covered the entire work area, totaling 100 samples for each layer. The spectral readings were obtained in the laboratory by the sensor FieldSpec 3 JR. in the range of 350 to 2500 nm. The spectral curves of the samples were separated into bands and heights, which were related to each attribute by multiple linear regression. The diffuse spectroradiometry proved to be more efficient in estimating sand, clay, sum of bases, and total iron, and less efficient in the estimation of organic matter and silt, in both horizons. The results obtained after validating the models showed that diffuse spectroradiometry is a promising tool, which can be used to determine soil properties.
reflectance; grid; multiple linear regression; prediction
