MODELLING SPATIAL VARIABILITY OF SOIL CHEMICAL ATTRIBUTES IN <em>TECTONA</em> <em>GRANDIS</em> STANDS IN CENTRAL-WEST BRAZIL
Keywords:
Geostatistics, teak, soil variability, precision forestry, nutritional deficienciesAbstract
The present study aimed to apply geostatistics to characterise spatial variability of chemical soil properties in a homogeneous teak stand. We plotted 91 samples in an area of 467 ha, sampling the soil at 0–20 cm depth for modelling spatial patterns of soil properties by semivariograms, interpolating and spatialising data by the ordinary point kriging method. Spatial variability of chemical soil properties varied in the area and can be precisely and efficiently modelled by geostatistics. We generated thematic maps for soil properties which were able to identify zones with nutritional deficiencies for teak, according to the recommended levels for the species. Evaluation of the site quality index according to the dominant height of trees revealed that the better site indices were in areas with higher concentrations of calcium and pH. Geostatistics proved to be suitable for the definition of homogeneous strata in teak stands. It can be used as an alternative method to regulate natural soil fertility in function of physiological requirements of the species, leading to a balanced input usage by forest producers and mitigation of environmental impacts.