Soil & Tillage Research 223 (2022) 105472 Available online 6 July 2022 0167-1987/© 2022 Elsevier B.V. All rights reserved. Fusion of Gamma-rays and portable X-ray fuorescence spectral data to measure extractable potassium in soils Said Nawar a , Florence Richard b , Anuar M. Kassim c , Yucel Tekin d , Abdul M. Mouazen b, * a Soil and Water Department, Faculty of Agriculture, Suez Canal University, Ismailia 41522, Egypt b Department of Environment, Faculty of Bioscience Engineering, Ghent University, Belgium c Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, 76100 Durian Tunggal, Melaka, Malaysia d Vocational School of Technical Sciences, Bursa Uludag University, Bursa, Turkey A R T I C L E INFO Keywords: Proximal soil sensing Data fusion Soil potassium Partial least squares regression ABSTRACT The detection and mapping of plant-extractable potassium (K a ) using proximal soil sensing and data fusion (DF) techniques are essential to optimise K 2 O fertiliser application, improve crop yield, and minimise environmental and fnancial costs. This work evaluates the potential of combined use of portable gamma ray and x-ray fuo- rescence spectroscopy for in feld detection and mapping of K a . After subjected to various pre-processing methods, spectral data were split into calibration (75%) and validation (25%) sets, and single sensor and DF models were developed using partial least squares regression (PLSR). Maps of K a were used to present spatial variability of potassium across an 8.4 ha Voor de Hoeves target feld, in Flanders, Belgium. Results showed that the gamma-ray PLSR model using wet soils had greater predictive ability with coeffcient of determination (R 2 ) = 0.71, ratio of performance deviation (RPD) = 1.89, root mean square error (RMSE) = 31.7 mg kg -1 , and ratio of performance to interquartile range (RPIQ) = 2.36 than the corresponding wet-XRF PLSR model (R 2 = 0.61, RPD = 1.64, RMSE = 48.8 mg kg -1 and RPIQ = 1.84). The DF PLSR model on wet soils, resulted in a more accurate K a predictive ability (R 2 = 0.75, RPD = 2.03, RMSE = 31.3 mg kg -1 and RPIQ = 2.79), compared to the individual gamma ray or XRF sensors in wet soils. The best accuracy was obtained with XRF spectrometer using dry samples (R 2 = 0.77, RPD = 2.14, RMSE = 26.5 mg kg -1 and RPIQ = 3.39). All K a prediction maps showed spatial sim- ilarity to the corresponding measured maps in the target feld. In conclusion, since DF increased the K a prediction accuracy compared to the single sensor models using wet soils, it is recommended to be adopted in future studies as a potential solution for K a measurement, mapping, and ultimately for site-specifc K 2 O fertilisation manage- ment. The XRF analysis of dry spectra is recommended as the best method for accurate measurement of K a . 1. Introduction Potassium (K) in soil, is an essential macronutrient for plant growth, and the second most abundant nutrient in plant tissues (Qiu et al., 2014; Sardans and Pe˜ nuelas, 2015). In soil, K exists in exchangeable and non-exchangeable forms. Non-exchangeable K forms are attached to a crystal lattice of mica or between layers of phyllosilicates (Sardans and Pe˜ nuelas, 2015; Z¨ orb et al., 2014). Exchangeable K consists of potassium adsorbed onto clay and organic matter particles, potassium dissolved in water as free ions, and plant extractable potassium (K a ). Potassium, like nitrogen and phosphorus, is a limiting factor in plant productivity, which can be managed by accurate potassium-based (K 2 O) fertiliser application (Sardans and Pe˜ nuelas, 2015). Precision agriculture (PA) requires accurate measurement of the spatial distribution of K a to enable site-specifc fertilisation that is ex- pected to sustain high crop yields and reduce environmental damage (Qiu et al., 2014). There is a range of proximal soil sensing technologies extractable for research and commercial purposes, including spectral and radiometric, electric and electromagnetic, acoustic, mechanical, pneumatic, and electrochemical methods (Kuang et al., 2012). Gamma-ray spectrometry is considered a relatively new and promising proximal soil sensing (PSS) for PA applications. It has been used for the measurement of various soil properties, e.g., clay content, pH, total ni- trogen, texture, and K a (Viscarra-Rossel et al., 2007; Heggemann et al., 2017). It was also reported to map variations in soil with high resolution when mounted on a vehicle (Van Egmond et al., 2010; Kuang et al., * Corresponding author. E-mail address: abdul.mouazen@ugent.be (A.M. Mouazen). Contents lists available at ScienceDirect Soil & Tillage Research journal homepage: www.elsevier.com/locate/still https://doi.org/10.1016/j.still.2022.105472 Received 31 December 2020; Received in revised form 26 May 2022; Accepted 29 June 2022