Geostatistical methods applied to soil fertility zoning

In conventional agricultural production systems, soil management is generally carried out without considering the spatial variability of its properties. This situation generates not only soil degradation but also an increase in production costs associated with the management of this factor. The objective of this research was to evaluate, through geostatistical methods, the spatial variability of soil fertility in Botana Experimental Farm of Universidad de Nariño. Spatial variability maps were estimated using the ArcGIS 10 program, the Kriging interpolation method, and the optimal ranges of soil fertility for the Andean region as projection parameters for making decisions related to land use. The fertility zoning of the farm was established, classifying soil as having high, medium, and low fertility. Most of the experimental farm had low fertility soils (20.7ha), and only 3ha had good conditions. Statistical analysis indicated a high variability in soil chemical properties. Properties such as pH and bulk density showed low variability.