Study On Estimating LAI And F_APAR From Hyperspectral Data In Crop Canopy

Canopy LAI and FAPAR are two important indicators to reflect growth status of crops population, they can quantitatively describe the original energy exchange structural information in the crop canopy and thy have important role in monitoring crop growth dynamic. They are highly related to vegetation growth vigor and yield. In this study, LAI and FAPAR of rice’s seedling stage, heading stage, maturity stage and cotton’s bud stage, flower stage, boll opening stage and the corresponding spectral reflectance was used to establish canopy LAI and FAPAR estimation model.By preprocessing the spectral data, we found that there was a apparent reflectance peak and absorption valley in the rice and cotton canopy spectral visible band, the green peak and red valley of each period were close to556nm and674nm respectively. Analysis the original spectral reflectance and first derivative spectral reflectance for each growth period of rice and cotton, we chose the green peak reflectance (Rg), red valley reflectance (Rr) and near-infrared reflectance (Rn) based on the original spectral reflectance, and the amplitude of the blue side (Db), red edge amplitude (Dr) based on the first-order differential spectral reflectance, and Rg/Rr,(Rg-Rr)/(Rg+Rr) based on the combined spectrum as the seven spectral characteristical parameters. Use these characteristical parameters, the canopy LAI and FAPAR of rice and cotton to do the regression analysis to establish characteristical parameters and estimation model. By evaluating the estimation model, we chose the best regression model of each growth period and tested them. This study explored the best estimation methods for LAI and FAPAR of each growth period of rice and cotton to provide the basis for crop field management and final production forecast.