Modelling nitrogen and phosphorus content at early growth stages in spring barley using hyperspectral line scanning

This paper addresses the possibilities for prediction of nitrogen and phosphorus content in barley plants, independently, using hyperspectral canopy reflections acquired under natural field conditions. A field trial consisting of spring barley was established and randomly applied with eight different levels of nitrogen (N) and phosphorus (P), respectively, in each plot. The canopy spectral reflectance data were obtained at three early growth stages under natural light conditions. The equipment used to perform measurements was a line scan spectrometer. Partial least square (PLS) regression was used on continuous spectra in the range of 400-750 nm and the total N and P contents were established through chemical analyses and used as references. The results showed that N content could be predicted with 81% accuracy and P content with 74% accuracy based on the canopy spectral reflectance. Furthermore, the regression models could be simplified by including the canopy growth stage information. The reflectance data could also be used to determine the growth stage with a certainty of 75%. (C) 2004 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd.