Diagnostic mapping of canopy nitrogen content in rice based on hyperspectral measurements

Timely assessment of canopy nitrogen content (CNC) in crops is critical for growth diagnosis and precision management of crops to generate higher yield and better quality while also minimizing adverse environmental impacts. The objective of this study was to determine the most suitable algorithm, using hyperspectral reflectance data, for the regional assessment of CNC at a critical growth stage in paddy rice. Ground-based hyperspectral datasets were obtained during the panicle formation stage under a wide range of plant and environmental conditions in Japan and China using spectroradiometers. A hyperspectral airborne dataset was obtained over a typical rice-growing region in Japan using the Compact Airborne Spectrographic Imager 3 (CASI-3). On the basis of a comprehensive analysis of the hyperspectral data, significant spectral indices (SIs) such as the normalized difference spectral index (NDSI) and ratio spectral index (RSI) were explored to provide an accurate and robust assessment of CNC. The capability of multivariable regression approaches, such as partial least-squares regression (PLSR) using the whole hyperspectral data or interval PLSR (IPLSR) using selected wavebands was also examined. Among various SIs, a simple index, RSI (D740, D522) using the first derivative (D) values at 740 nm and 522 nm, was found to be most accurate and robust for the assessment of CNC. The predictive ability of the index was comparable to those of PLSR and IPLSR. Independent validation using the airborne dataset supported the robust applicability of the new SI. The CNC was closely related to the conventional diagnostic indicators based on direct plant measurements. The results demonstrated the operational applicability of hyperspectral measurements for diagnostic mapping of CNC on a regional scale. The investigation based on the precise dataset for rice will be a good basis for remote sensing of canopy nitrogen content in a wide range of vegetation. (C) 2012 Elsevier Inc. All rights reserved.