Monitoring model of leaf area index of winter wheat based on hyperspectral reflectance at different growth stages
He Jia,Liu Bingfeng,Li Jun
DOI: https://doi.org/10.3969/j.issn.1002-6819.2014.24.017
2014-01-01
Abstract:Hyperspectral remote sensing can rapidly and nondestructively acquire vegetation canopy information. It is an important real time technology to monitor and manage crop growth. Leaf area index (LAI) is a key parameter for crop growth evaluation and yield prediction. The objectives of this study were to establish wheat LAI estimation model based on winter wheat (Triticum aestivum L.) canopy hyperspectral reflectance with different rates of nitrogen or phosphorus application, and to improve the forecast precision of the LAI estimation model at different growth stages of winter wheat on the Loess Plateau of China. The experiments were carried out during 2009-2014 at Northwest A&F University, Yangling, China. The treatments included different winter wheat varieties with various drought resistance grown under five nitrogen fertilizer application rates (0, 75, 150, 225 and 300 kg N/hm2) and four phosphorus application rates (0, 60, 120 and 180 kg P2O5/hm2) LAI and canopy hyperstpectral reflectance of different varieties under fertilizer treatments were measured at jointing, booting, heading, grain filling and maturity stage, respectively. Then LAI monitoring models at different growth stages of winter wheat were constructed by using correlation analysis, regression analysis. The results showed that LAI of wheat was increased with increase in nitrogen and phosphorus application rate at different growth stages, and LAI from jointing to maturity showed a parabolic curve, and the maximum LAI of wheat was at heading stage. When nitrogen or phosphorus application was sufficient, the canopy hyperspectral reflectance of wheat was reduced 3%-5% in the visible wavelength (P<0.05), and increased 4%-10% in the near infrared wavelength (P<0.05). There were significant (P<0.01) correlation between Optimized Soil Adjusted Vegetation Index (OSAVI), Enhanced Vegetation Index Ⅱ (EVI2), New Vegetation Index (NVI), Modified Normalized Difference Vegetation Index (mNDVI), and Modified Simple Ratio Index (mSRI) to LAI, the range of correlation coefficient were from 0.852 to 0.987 at different growth stages. Monitoring models based on OSAVI, EVI2, EVI2, mNDVI and mSRI produced better estimation for LAI at jointing, booting, heading, grain filling and maturity respectively, and R2 were respectively 0.952, 0.979, 0.989, 0.960 and 0.993 (P<0.01). Meanwhile, compared the predicted value and measured value to verify reliability and applicability of the model, relative error of the measured value and predicted value were 13.0%, 13.5%, 12.8%, 12.6% and 0.14.0%, and root mean square error were 0.313, 0.336, 0.316, 0.316 and 0.324, at jointing, booting, heading, grain filling and maturity stage, respectively. Therefore, vegetation indices of OSAVI, EVI2, EVI2, mNDVI and mSRI was the most suitable indeces for monitoring winter wheat LAI at jointing, booting, heading, grain filling and maturity, respectively. There was high prediction precision with different vegetation indices monitoring LAI of winter wheat at different growth stages. These conclusions has important implications for monitoring LAI of winter wheat in large area on the Loess Plateau. Meanwhile, there is a high prediction accuracy of monitoring model based on the different vegetation indices at different growth stages of winter wheat.This result provides technical support for growth monitoring of winter wheat at different fertility and different growth stages for farmers..