External Parameter Orthogonalization (EPO) for Soil Moisture Correction of Eastern European Visnir-DRS Data.

As developing countries such as those of Eastern Europe look to advance their agricultural markets through large scale agronomic production, high resolution mapping of soil resources will be essential. Visible near infrared diffuse reflectance spectroscopy (VisNIR DRS) is a proximal sensing technique which provides quantitative data germane to physicochemical soil properties in seconds. However in-situ, variable soil moisture confounds VisNIR spectroscopy for predictions of soil properties. Though the external parameter orthogonalization (EPO) algorithm has shown promise in removing the effect of soil moisture on soil spectra of field moist soils, it has not been tested in Eastern European soils. Further, physical integrations between soil minerals and soil water are mineral dependent causing uncertainty as to the effectiveness of EPOs in new geomorphological settings. Thus, the objective of this work was to test the effectiveness of EPO on intact field moist soil spectra and a dried-ground spectral library from highly variable soils from Romania and Bulgaria, and to use this diverse dataset to assess the sensitivity of the EPO-PLS parameterization and performance to enhance the calibration spectral dataset. A dried-ground spectral library of 500 soils collected from Eastern Europe was transformed using the EPO P-matrix from 150 surface soils collected independently of the library and scanned as field-moist intact and dried-ground conditions. Results show that EPO can remove the effect of soil water from field-moist spectra for soils in this area. Bootstrapping analysis was performed to assess the sensitivity of the EPO-PLS procedure to changes in the structure of the calibration spectral dataset. While research has shown that EPO is effective on smectitic and kaolinitic soils, further research on mixed mineralogy is needed to understand its impacts on EPO effectiveness.