Inversion Of Winter Wheat Leaf Area Index And Nitrogen Vertical Distribution With Hyperspectral Data

Leaf area index (LAI) and nitrogen content are not only the important parameters to estimate crop yield, but also indispensable parameters to monitor crop growth status and implement the field management. Traditional remote sensing inversion mainly focuses on LAI and nitrogen content of crop surface layer, but LAI and nitrogen vertical-layer distribution are uneven. When under the condition of crop nutrient deficient, deficiencies usually exhibit in the bottom layer leaves and production need early and timely detection of crop nutrient vertical distribution. In this study, LAI and foliage nitrogen vertical-layer distribution inversion methods were studied based on the data of winter wheat LAI, the vertical distribution of foliage nitrogen and ground hyperspectral data for years.In this paper, the main content and results are as follow:(1) The LAI of winter wheat obtained from Beijing city had been estimated successfully by support vector machine regression (SVR) model built with LAI and wavelet coefficients of hyperspectral reflectance. The inversion results of this paper method and other five methods, such as selected vegetation indices and partial least-square (PLS) regression models, were analyzed. It was found that the sensitive bands to assess LAI were680nm,739nm,802nm, and895nm, and the corresponding wavelet decomposition scales were8,4,9, and8determined by continuous wavelet transform (CWT), respectively. The decision coefficient (R2) of regression equation between LAI and wavelet coefficient was significantly higher than that of between LAI and canopy reflectance. The SVR model based on wavelet coefficients performed best with R2of0.86, and RMSE of0.43, while the regression models based on two common spectral vegetation indices (NDVI and RVI) performed poor in estimating LAI of winter wheat’s multiple birth period(R2<0.76,RMSE>0.56). The combination of CWT and SVR is feasible to realize remote sensing inversion of LAI in the whole growth period of winter wheat.(2) Considering the effect of NIR/red ratio on the reflection of canopy and ground parameters, two new spectral vegetation indices, normalized difference ratio index (NDRI) and enhanced ratio vegetation index (ERVI), have been improved from normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). The results showed that:(a) the LAI estimation models by NDVI or EVI should be established separately for winter wheat with different canopy geometric structures;(b) the NDRI and ERVI at view zenith angle (VZA) of40°had the highest accuracy for estimating LAI in winter wheat with different crop geometric characteristics, comparison with other commonly used spectral vegetation indices (e.g., NDVI) or the values from other view angles;(c) the NIR/red ratio at VZA of40°an represent canopy cover and crown shape in the canopy geometry.(3)This study presents the method that partial least square regression (PLSR) was used to extract foliage nitrogen vertical-layer distribution by exploring multi-angle spectrum, and it will help to solve the problem of early and timely diagnosis of crop nutrient status. The results indicated that:a) the canopy reflectance at nadir,40°and50°; at nadir,30°and40°; and at nadir,20°and30°were selected to estimate foliage nitrogen density (FND) at upper layer, middle layer and bottom layer, respectively, b) For each layer, three PLSR analysis models with FND as the dependent variable and three types of vegetation indices (nitrogen reflectance index (NRI), normalized pigment chlorophyll index (NPCI) or the combination of NRI and NPCI) at corresponding angles at corresponding angles as the explicative variables were established, and independent model verification showed that the PLSR analysis models with the combination of NRI and NPCI at corresponding angles as the explicative variables were most accurate in estimating FND for each layer by taking more bands into consideration.The result of this study provide theoretical basis for early and timely diagnosis of crop growth and nutrient status, and also provide the reference for reasonable setting of bands and angles for a new generation of remote sensing sensor.