Measuring and mapping potassium in agricultural fields using gamma spectroscopy

Authors: Kavetskiy, Aleksandr; YAKUBOVA, GALINA - AUBURN UNIVERSITY; SARGSYAN, N - AUBURN UNIVERSITY; Prior, Stephen - Steve; Torbert, Henry - Allen; CHIN, B - AUBURN UNIVERSITY

Submitted to: Inst of Electrical and Electronic Engineers Transactions of Nuclear Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/30/2021
Publication Date: 10/19/2021
Citation: Kavetskiy, A.G., Yakubova, G., Sargsyan, N., Prior, S.A., Torbert III, H.A., Chin, B.A. 2021. Measuring and mapping potassium in agricultural fields using gamma spectroscopy. Inst of Electrical and Electronic Engineers Transactions of Nuclear Science. 68(10):2550-2558. https://doi.org/10.1109/TNS.2021.3109486.
DOI: https://doi.org/10.1109/TNS.2021.3109486

Interpretive Summary: Potassium plays significant roles in enhancing crop quality and is essential for proper growth. While the amount of readily available soil K is most important for crop responses, correlation between readily available and total soil K can be useful for developing fertilizer programs for agricultural fields. This project covers the development of an alternative way of measuring total soil K. This method paired GPS with gamma spectra measurements (i.e., scanning mode) to map total soil K in agricultural fields. Created maps were comparable to known soil K distribution data and demonstrated the workability of this method for measuring total soil K.

Technical Abstract: The gamma peak with a centroid at 1.46 MeV associated with the natural decay of the 40K isotope was used to measure total soil K. A gamma spectrometer in scanning mode (NaI detectors, total volume 7.5 dm3) paired with a GPS device was used to measure and map total K content in agricultural fields. A measurement algorithm and software for system operation, data acquisition, data processing, and map creation (using ArcMap) was developed and reported. The calibration coefficient for K content determinations in real soils was defined by measuring small sample volumes of known K content and by Monte-Carlo simulation (Geant4) of the gamma spectra for different sample volumes to semi infinity. The workability of the developed Monte-Carlo simulation code was validated by agreement between simulated and measured spectra of a volumetric K sample at different sample-to-detector distance. Two reproducible maps of an 8.5 ha field were created, each based on ~ 2.5 hours of scanning. Comparing the created maps to known soil K distribution data demonstrated the workability and applicability of the described method for total soil K measurements.