Coupling EMI and NIR spectroscopy for soil mapping with limited number of samples

Digital agriculture needs accurate and detailed maps of soil features without excessive increasing of costs for soil analysis. Proximal soil sensing, in particular electromagnetic induction (EMI), is increasingly being used in the market of professional users and consultants to provide high detailed maps of electrical resistivity (or conductivity) related to several soil features. Near Infrared (NIR) spectroscopy is also a rapid and low-cost technique to predict soil features, partially replacing the conventional laboratory analysis. The objective of this work is to test a protocol for soil mapping on a small farm scale (9 ha), using both EMI and NIR spectroscopy, and a minimum set of soil samples analyzed by traditional laboratory analysis. Using only 5 analyzed local samples and an external spectral library, maps of soil organic carbon, clay, and sand, were obtained with fair accuracy.