Hyperspectral characteristic analysis of a developing cotton canopy under different nitrogen treatmentsDehua Zhaoa, Jianlong Lia and Jiaguo Qib
a Department of Biological Science and Technology, Nanjing University, Nanjing 210093, China
b Center for Global Change and Earth Observations, Michigan State University, East Lansing, MI 48823, USA
(Received 21 August 2003; accepted 18 May 2004)
Abstract - The objective of this study was to analyze the differences of crop agronomy parameters, canopy spectral reflectance and spectral indices induced by different nitrogen application rates, and to assess the potential of discriminant analysis in discriminating cotton canopies at different nitrogen treatments. We conducted an experiment in cotton fields treated with four nitrogen application rates: 0%, 50%, 100% and 200% of the recommended rate. Results suggested that no single spectral index or canopy variable can fully characterize the canopies nitrogen status throughout the growing period. With a single vegetation index, discriminant analysis provided less than 45%, 48.8%, 61.6% accurate classifications of cotton canopies by nitrogen treatment during early, mid and late season, respectively. Nevertheless, using multi-vegetation-index model by stepwise procedure, 74.4%, 83.1% and 89.6% accuracies could be obtained during early, middle and late season, respectively.
Key words: cotton / hyperspectral remote sensing / spectral indices / nitrogen application rates / leaf area index / chlorophyll
Corresponding author: Jianlong Li email@example.com
© INRA, EDP Sciences 2004