Spatial patterns of magnetic susceptibility optimized by anisotropic correction in different Alisols in southern Amazonas, Brazil

• Autor(a) principal: Brito, Wildson Benedito Mendes
• Data de Publicação: 2022
• Outros Autores: Campos, Milton César Costa, de Souza, Fernando Gomes, Silva, Laércio Santos [UNESP], da Cunha, José Maurício, de Lima, Alan Ferreira Leite, Martins, Thalita Silva, de Oliveira, Flávio Pereira, de Oliveira, Ivanildo Amorim
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1007/s11119-021-09843-6; http://hdl.handle.net/11449/229424
• topic: Agricultural conversion; Isotropy; Kriging; Soil magnetism; Spatial variability

Changes in primary cover for agricultural crops in Amazonas region influence the phenomenon of spatial variability in soil properties. This phenomenon is still studied assuming that the spatial data is isotropic, but does not consider the anisotropic pattern of soil properties. Thus, the aim of this work was to characterize, identify and correct isotropic patterns of magnetic susceptibility (MS) measurements using anisotropic models that actually represent the spatial aspects of the data. Three cultivation areas and one under native forest, classified as Haplic Alisol, were georeferenced and sampled by a mesh system covering 192 samples per area. Texture, X-ray diffraction and frequency-dependent (χfd) and mass-specific (χlf and χhf) magnetic susceptibility analyzes were performed. Then, classical and geostatistical analyzes were applied to the data, assuming isotropy and anisotropy. All χ frequencies were shown to be spatially dependent, satisfying the isotropy hypothesis. Thereby, the application of anisotropic analysis was able to confirm the presence of all types of anisotropy in Alisols. Anisotropic correction provided an improvement in models that fit the directional trends within the areas, and provided a reduction in the nugget effect and an increase in the correlation ranges. Thus, the generated kriging maps improved the patches of zonal trends of greater or lesser χ that stand out at the level of sub-regions. These zones should, therefore, be used as indicators of variability, paying special attention during their management, especially in research related to the delimitation of specific management zones.