Spatial Analysis of Granulometry and Humic Substances of an Ecological Restoration Area in the Brazilian Atlantic Forest

Abstract

The present study aimed at spatially analyzing granulometry and humic substances (HS) of an ecological restoration area in the Brazilian Atlantic Forest. In the study area, 49 points were systematically distributed, and soil was collected at each point, at a depth of 0-10 cm. The soil attributes evaluated were: total clay, total sand, silt, and organic carbon in the fulvic acid fraction (C-FAF), humic acid fraction (C-HAF), and humic fraction (C-HUM). Semivariogram analysis was performed for each variable, with spherical, exponential, and Gaussian semivariogram models tested. In order to choose the best model, the following parameters were used: Akaike Information Criterion (AIC), determination coefficient (R²), root mean square error (RMSE), Willmott concordance index (d), and degree of spatial dependence (DSD %). Mapping was performed using the Ordinary Kriging method. The attributes showed spatial dependence. The spherical model was adjusted for all variables. From spatialization, the attributes, except clay, showed a heterogeneous distribution, which may be due to the high range value. The distribution of soil texture depended mainly on soil class and relief. A correlation was observed in the distribution of the silt and C-HUM attributes. The ecological restoration area had HS content values close to those reported in studies conducted under conditions similar to those in the present study. Therefore, it appears that the study area is probably undergoing recovery; however, a thorough and continuous study of the area is essential for confirmation.