Soil fertility analysis by validation and Kriging interpolation of soil parameters

Soil fertility is one of the most important elements in crop nutrition and its spatial variation can be determined through geostatistical techniques that allow mapping and delimiting management areas. One of the important advantages of geographic information systems (GIS) is spatial analysis, particularly the use of interpolations of different soil physicochemical variables. Thus, the objective of this research is to analyze soil fertility in the community of Santo Domingo, Huasca de Ocampo, Hidalgo, using thematic maps created by the ordinary Kriging interpolation method and validated with field and cross validation techniques for the soil fertility variables: nutrient content N, P, K, Ca, Mg and Na, and the properties pH, EC, MO and CIC. The statistical information for the variables proved to be uniform and easily predictable with the exception of pH and OM where little representativeness of the mean was observed associated with intrazonal variability. The model selected for the adjustment of the experimental semi-variogram, which best adjusted the variables studied, was the Gaussian model, except for the EC variable that was adjusted to the spherical model. The results show soils with low pH, which indicates acid soils, therefore, Ca, P and Mg nutrients are less available to the plant. Likewise, primary nutrients, such as N, P and K are found in deficient amounts. In conclusion, the maps obtained in this study from the ordinary Kriging model and the two types of validation used can be a useful tool as an approximation and reference to determine to a good extent the spatial distribution and variability of soil fertility properties. The values of MO, N, K, P and pH mainly defined the current state of soil fertility and evidence its degradation.