Monitoring Nitrogen Status And Growth Characters With Canopy Spectral Reflectance In Summer Maize

The summer maize is the most main grain crop of the summer crops in Huanghuaihai region. Real-time monitoring crop growth conditions and nitrogen status through spectral techniques would provide the theoretical basis and key techniques for non-destructive monitoring in precision crop management of maize-growing regions in Huanghuaihai region. In this study, the maize of Huanghuaihai region on wheat stubble field were selected as our interesting object. Based on the technology of canopy reflectance spectral and the test of physiological index, the characteristics of canopy spectral reflectance under different nitrogen status and their correlation to LAI, biomass, leaf nitrogen status, leaf soluble sugar to N ratio and leaf pigment contents in summer maize were analyzed. The sensitive spectrum parameters and quantitive regression models for nitrogen status and growth characters were established.The main results were as follows:1. Based on the change patterns of leaf area index and biomass under different nitrogen rates and growth stages, the relationships of leaf area index and biomass to canopy spectral reflectance and spectral parameters were investigated, and sensitive spectral parameters and quantitative equations for predicting LAI, dry leaf weight and dry aboveground biomass were established. We found that LAI was correlated well with normalized difference vegetation index, ratio vegetation index from 730 nm and near infrared bands(780～1050 rm), especially NDVI(850,730) correlated the best. Spectral parameter NDVI(850,730) could be used to deduce LAI with linear model in precise. Testing the monitoring models with independent dataset indicated that the spectral parameter of NDVI(850,730) could give an accurate estimation. Dry leaf weight and dry aboveground biomass can be predicted using the same spectral parameters. Dry leaf weight and dry aboveground biomass were correlated well with normalized difference vegetation index, ratio vegetation index from 550 nm and near infrared bands(780,900,950 nm), and NDVI(900,550) gave the optimal correlation. Spectral parameter NDVI(900,550) could be used to deduce dry leaf weight with exponential model in precise. Testing the monitoring models with independent dataset indicated that the spectral parameter of NDVI(900,550) could give an accurate estimation. 2. Based on the technique of spectra analysis, many characteristic bands and derived spectral parameters were obtained. The quantitative relationships between leaf nitrogen status and canopy reflectance spectra, and the sensitive parameters and monitoring equations of leaf nitrogen accumulation were put forward. Analyzing individual growth stage showed that there were highly significant correlation between leaf nitrogen content and canopy vegetation index NDVI (800,550), NDVI (850,550), RVI (800,550) and RVI (850,550) except the milky stage, and leaf nitrogen content could be predicted using the same vegetation index for the whole growth stages. Leaf nitrogen accumulation can be fitted using the same vegetation index in the whole growth stages. It was found that leaf nitrogen accumulation was fitted well with normalized difference vegetation index, ratio vegetation index from 550nm and near infrared bands(780,800,850,900,950,1050nm), MSR, OSAVI, MSAVI and SAVI. The result that testing the monitoring models with independent datasets indicated that OSAVI was the best parameter to predict leaf nitrogen accumulation.3. The regression analysis between spectral parameters and leaf soluble sugar to N ratio indicated that quantitative models for monitoring soluble sugar to nitrogen ratio were investigated in summer maize. The proper time for monitoring leaf soluble sugar to nitrogen ratio should be from booting to silking. Comprehensive analysis correlation between leaf sugar nitrogen ratio and spectral parameters of booting, tasseling and silking periods of two varieties were done and 13 better performance spectral parameters were selected. Testing the monitoring models with independent dataset indicated that the linear model of spectral parameter TVI were the best model to estimate leaf soluble sugar to nitrogen ratio.4. The regression analysis between spectral parameters and different pigment forms in leaves indicated that quantitative models for monitoring chlorophyll a, chlorophyll b, chlorophyll a+b and and carotenoid were investigated in summer maize. Different pigment forms in leaves can be predicted using the same spectral parameters of NDVI(950,550). The chlorophyll a and chlorophyll a+b contents could be more reliably estimated than chlorophyll b and carotenoid.