Spatial variability of organic fractions of soils cultivated with banana BRS Princesa and Tifton 85 in the definition of management zones
Spatial dependences, geostatistics, humic substancesThe aim of this study was to evaluate the spatial variability of chemical and physical parameters and fractions of organic matter in soils cultivated with Tifton 85 and banana BRS Princesa and characterize them in management zones. The experiments were carried out in areas located at the Federal Rural University of Rio de Janeiro, Seropédica-RJ, in cultivation of Tifton 85 at Fazenda Feno Rio and in the Horticulture Sector in organic cultivation of banana BRS Princesa. For the Tifton 85 area, 50 points were collected in irregular mesh and deformed soil samples were taken at depths of 0-0.20 m and 0.20-0.40 m. Data on the sum of exchangeable bases (SB), total cation exchange capacity (T value), base saturation (V%), organic matter (MO), total clay, total sand, silt, carbon in the fuvic acid fraction (C-FAF), humic acid fraction (C-FAH), humine fraction (C-HUM), particulate carbon (COp), mineral-associated carbon (COam), light organic matter (MOL, content of C and N), estimation of fresh mass (MF) of Tifton 85 and the need for liming (NC) at a fixed and variable rate for a depth of 0-0.20 m. For the banana area, 60 georeferenced points were collected and samples of deformed soil at 0-0.05, 0.05-0.10 m, 0.10-0.20 m and 0.20-0 were collected , 40 m. Soil attributes were analyzed pH, Ca, Mg, Al, Na, K, P, H+Al, SB, T value, V%, fractions of total sand, total clay and silt, total organic carbon (COT ), COp and COam (at 0-0.05 m, 0.05-0.10 and 0.10-0.20 m), soil resistance to penetration (RSP) and gravimetric soil moisture. The RSP was obtained from the cone index (CI) using an impact penetrometer and the CI was measured to a depth of 0.40 m. The data were submitted to exploratory analysis and descriptive statistics, followed by the geostatistical study in which each variable was subjected to modeling of theoretical semivariogram and ordinary kriging, obtaining spatial variability maps. Multivariate analysis was performed to design the management zones (ZM) based on the principal component analysis (PCA) followed by the k-means method. In the area cultivated with Tifton 85, the spatial distribution of fractions of organic matter in the soil suggests higher levels in regions with a high probability of occurrence of higher crop yields. In the organic cultivation of the BRS Princesa banana, the spatial relationships of nutrients available in the soil indicate a nutritional imbalance. The RSP values evaluated in the bananal suggest greater compaction in the subsurface. If the banana trees are damaged and management must be carried out to improve the edaphic conditions of the banana plantation, such as subsoiling located at a different depth. It was possible to generate two ZM for the different experimental areas for the purpose of fertilization recommendation. It was found that the localized management with the application of organic fertilizer can be indicated, and higher doses should be applied in one MZ when compared to the other, given the greater need for increasing organic matter in the soil in specific places in the experimental areas.