External parameter orthogonalisation of Eastern European VisNIR-DRS soil spectra

The quality and precision of spatial soil information varies tremendously worldwide, owing to differential access to technology, laboratory facilities, and adequate funding. Some countries offer detailed soil maps, but in developing countries such as Romania, Bulgaria, and others from Eastern Europe, such data are limited. Visible near infrared diffuse reflectance spectroscopy (VisNIR DRS) is a proximal and in situ sensing approach, which can quickly provide quantitative data reflecting multiple soil physicochemical properties. If in situ VisNIR DRS is shown effective in Eastern Europe, it would greatly enhance the collection of high quality soil data in support of mapping, environmental modeling, soil and water resource assessment, and precision agriculture. The present study was conducted on a soil library (500 samples) collected from mineral surface soils of Romania and Bulgaria. The samples covered a wide range of soil types and land use. VisNIR spectra were acquired on all samples with a portable spectroradiometer and then soil mineralogy was established via standard laboratory methods. The entire dataset was split into three subsets for two different external parameter orthogonalization (EPO) approaches; using a combination of field-moist intact cores and processed samples (EPO1) and only processed samples (EPO2) to remove the moisture signature from moist spectra of both moist intact and ground samples to improve clay predictions. Results indicated that using EPO1 and EPO2 it was possible to remove 5% and 29% of the error introduced by the wet/in situ measurements, respectively. However, the complex interactions between both soil moisture and mineral composition on the spectra possibly prevented EPO from fully removing the in situ effect on spectra when applied to a sample space with high variability in mineral composition. In situ VisNIR DRS shows important advantages over traditional laboratory approaches, allowing for more analytical density and less expensive soil information, with minimal reduction in data quality. The use of in situ VisNIR DRS with the EPO has excellent potential for providing data needed to support optimized soil resource mapping and land management decisions in Eastern Europe.