01112nam a2200157 a 450000100080000000500110000800800410001910000220006024500920008226001780017452005130035265000250086570000210089070000250091170000180093621150642020-01-03       2019    bl uuuu     u00u1 u    #d1 aVASQUES, G. de M.  aPredicting soil clay content from NIR, gamma-ray and XRF curves.h[electronic resource]  aIn: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: proceedings... Viçosa, MG: SBCS, 2019. v. 2, p. 535-536. WCSS 2018.c2018  aIn this study, data from NIR, gamma ray and XRF curves, and three multivariate methods (partial least squares regression - PLS, random forest - RF, and support vector machine - SVM) were used to predict soil clay content at 0-10-cm depth. Training and validation data included 103 and 25 samples, respectively. Gamma ray and XRF data were taken in situ at the soil surface, using portable sensors, whereas NIR reflectance curves (800-2500 nm) were measured from airdried fine earth samples in the laboratory.  aSensoriamento Remoto1 aRODRIGUES, H. M.1 aTAVARES, S. R. de L.1 aCOELHO, M. R.