Dissertations and Theses

Characterization of specific management areas in the context of soil-relief relations.

Author: José Marques Júnior

Keywords: soil attributes; plant attributes; pedometry; geostatistics; magnetic susceptibility; diffuse reflectance; multivariate analysis

Summary

The objectives of this work are: a) To understand the spatial relationships between the physical and chemical attributes of soil and relief, and to evaluate its influence on the production and qualities of citrus and sugarcane; b) Use of diffuse reflectance spectroscopy and Magnetic Susceptibility techniques to evaluate soil physical and chemical attributes, in the context of the soil-relief relationship; c) Identification of specific areas of management, using soil attributes, diffuse reflectance spectroscopy and magnetic susceptibility, and their relationship with relief. Two experiments were performed. In the first one, samples were collected in a regular interval of 50 meters, at a depth of 0.00-0.20 m, to determine soil attributes: clay content, organic matter content, water content, soil stability aggregate, macropores, micropores, total pore volume, saturated soil hydraulic conductivity, soil density and soil penetration resistance in a dystrophic Red Latosol under orange pear cultivation. The attributes of the citrus fruit analyzed were: total soluble solids, titratable total acidity, ratio, yield, juice yield and fruit size, at three different times. Data were analyzed by descriptive statistics, geostatistics and canonical correlation analysis (CCA). The results of the geostatistical analysis showed that the spatial variability of soil and fruit attributes is influenced by the relief forms. The temporal variability of fruit attributes is also influenced by the shapes of the relief, conditioning a differentiated ripening gradient for each relief compartment. The canonical variables of the first canonical pair had in common 77% of the variance explained. The relief forms were effective in the identification of homogeneous areas of management. In the second experiment, soil samples were collected in a 10-meter-wide mesh at a depth of 0.00-0.20 m in an area of ​​1 ha in an eutrophic Red-Yellow Argisol under cane- of sugar. The soil attributes were: clay content, total pore volume, soil bulk density, average weighted aggregate diameter, penetration resistance, water content, available P, adsorbed P, K, Ca, Mg, H + Al, organic carbon, pH, cation exchange capacity, magnetic susceptibility and the soil spectrum in the visible and near infrared range. Sampling of sugarcane plants was done at the same collection points as the soil. In the plant, the following attributes were evaluated: productivity, Brix, broth pol, percentage of fiber and total reducing sugars. Data were analyzed using descriptive statistics, multivariate statistics and geostatistics. The results showed significant correlations of soil and plant attributes with magnetic susceptibility and spectrum bands. In this way, the results allow to conclude that diffuse reflectance measurements, especially in the near infrared range, and magnetic susceptibility measurements can be used to predict other soil attributes and to identify management zones for sugarcane.