Soil spatial variability mapping and optimized sampling supported by sensing techniques: bases for a more efficient and sustainable precision agriculture

Abstract

Agricultural fields are not uniform in their extension, and, therefore, the adoption of precision agriculture (PA) is mandatory to obtain greater efficiency and sustainability of agricultural production. To make this approach viable, sensing techniques to support the mapping of soil properties are essential. The mapping of soil chemical fertility from systematic sampling for variable-rate fertilization is the most adopted practice worldwide. However, several other soil properties impact crop development and productivity, which have been neglected in PA research. Therefore, the main goal of this research is to generate knowledge for the development of protocols for the spatial diagnosis of soil at the crop level so that PA can be carried out more holistically, allowing a general understanding of soil quality, its natural spatial variability, and also the portion of the variability generated by the anthropic influence that demands site-specific management. Therefore, this research project aims to bring pedology closer to PA, developing innovative knowledge both in scientific and practical terms. The proposal is divided into three main scientific objectives (sensing techniques, sampling optimization and multivariate prediction, and soil mapping approaches) and will feature long-term monitoring of three commercial fields that represent a good part of the national cropping systems (sugarcane, grains, and crop-livestock integration). This document also presents a research and teaching integrated plan resulting from the project's development. (AU)