Potential of apparent soil electrical conductivity to describe the soil pH and improve lime application in a clayey soil

- Apparent electrical conductivity as a powerful tool to describe soil variability.
- Improved fertiliser applications by a feasible soil sampling.
- Sustainable production by a site-specific management according soil requirements.

One of the main limitations in applying fertilisers at a variable rate according to soil requirements is the number of samples required to spatially represent soil attributes. Currently, to map soil attributes with high accuracy, a high sample density is needed, which is often unfeasible. One technique that minimizes the number of samples required is the use of maps that have prior information of the soil spatial variability, making it possible to identify representative sampling locations in the field. Soil apparent electrical conductivity (ECa) is a powerful indicator of the characteristics of the soil matrix, and it can be easily measured. The objectives of this study were to determine the efficacy of ECa for mapping the spatial variability of soil properties using a targeted and reduced number of samples (which is economically feasible for growers). A high sampling density was applied in an experimental sugarcane field to assess the spatial variability of soil attributes, and ECa was read by a direct contact sensor. Thematic maps of important agronomic attributes were generated using the high sample density and compare with a targeted and reduced sampling helped by ECa measurements. The methodology used to obtain the spatial variability maps of soil chemical and physical properties indicated that it was possible to obtain maps of acceptable accuracy (r = 0.82 and 0.59 for clay and pH, respectively) that could be used to formulate lime recommendations for variable-rate applications. The total lime spent (28 tons) based on the application maps generated by a targeted sampling mesh (20 sampling points) was very similar to that (30 tons) obtained based on high-density sampling (204 sampling points), representing a large improvement in soil sampling methodology. Here we demonstrated that ECa can improve prediction maps and enables efficient sample orientation. This approach provides an interesting way to map soil properties over large areas using a feasible sampling method and can help farmers in crop management while respecting environmental and soil requirements.

285