Aims Biomass and leaf area index (LAI) are important parameters for indicating crop growth potential and photosynthetic productivity in wheat. Non-destructive, quick assessment of leaf dry weight and LAI is necessary for growth diagnosis and cultural regulation in wheat production. The objectives of this study were to determine the relationships of leaf dry weight and LAI to ground-based canopy hyperspectral reflectance and spectral parameters and to derive regression equations for monitoring leaf dry weight and LAI in winter wheat (Triticum aestivum) with hyperspectral remote sensing.
Methods Three field experiments were conducted with different wheat varieties and nitrogen levels for three growing seasons, and time-course measurements were taken on canopy hyperspectral reflectance and leaf dry weight and LAI during the experiments. Experiment one was conducted in 2005-2006 to construct a monitoring model with four N rates of 0, 90, 180 and 270 kg·hm-2 using cultivars ‘Ningmai9’ and ‘Yumai34’ (low and high protein types, respectively). Experiment two was undertaken in 2004-2005 to construct a monitoring model with four N rates of 0, 75, 150 and 225 kg·hm-2 using cultivars ‘Ningmai9’, ‘Yangmai12’ and ‘Yumai34’ (low, medium, high protein types, respectively). Experiment three was conducted in 2003-2004 to test a monitoring model with four N rates of 0, 75, 150, 225 and 300 kg·hm-2 using cultivars ‘Ningmai9’, ‘Huaimai20’ and ‘Xuzhou26’ (low, medium, high protein types, respectively).
Important findings Leaf dry weight and LAI in wheat increased with increasing nitrogen rates and with significant differences between stages of growth. The dynamics of leaf dry weight and LAI during growth exhibited single peak patterns. The sensitive spectral bands were located mostly within red light and near infrared regions, with correlation coefficients <-0.60 in 590~710 nm and >0.69 in 745~1 130 nm. The regression analyses between existing vegetation indices and leaf dry weight and LAI revealed that some key spectral parameters could accurately estimate changes in leaf dry weight and LAI across a broad range of stages of growth, nitrogen levels and growing seasons, with unified spectral parameters for each growth parameter. Among them, regression models based on RVI (810, 560), FD755, GMI, SARVI (MSS) and TC3 produced better estimation of leaf dry weight and LAI. Testing of the monitoring models with an independent dataset indicated that the spectral indices of RVI (810, 560), GMI, SARVI (MSS), PSSRb, (R750-800/R695-740)-1, VOG2 and mSR705 gave accurate growth estimation under the experimental conditions. Overall, leaf dry weight and LAI in wheat could be monitored by key vegetation indices, with more reliable estimation from RVI (810, 560), GMI and SARVI (MSS).