The sudden increase in vegetation canopy reflectance from low reflectance in red band to near infrared band forms the red-edge spectral characteristics, which is unique to healthy vegetation. Many parameters that can describe this characteristic have been designed and developed as important indicators of crop growth and nutrition status. However, few studies have systematically compared and evaluated the applicability of these red-edge parameters to estimate winter wheat leaf CHL values at different growth stages and nitrogen application levels. In this study, canopy spectral reflectance and leaf CHL of winter wheat at 4 key growth stages (jointing stage, heading stage, flowering stage and filling stage) and 3 nitrogen application levels were obtained through a 4-year field experiment. The sensitivity of 47 spectral red-edge parameters to CHL was evaluated, and the relative importance of spectral red-edge parameters was used to optimize the random forest machine learning model to estimate winter wheat CHL. The results showed that the sensitivity of spectral red edge parameter to CHL was affected by the growth period and nitrogen application level of winter wheat, and the correlation R² between the best red edge parameter and CHL in a single growth period was between 0.39 and 0.89. The best red-edge parameter in the whole growth period was NDDRmid, and the coefficient of determination between it and CHL was 0.76. The sensitivity was the highest in filling stage, and the coefficient of determination between the best red edge parameter RVI5 and CHL was 0.89 and the R² between red edge parameters REPRpi, NDDRmid, RVI2, RVI4, RVI5, RVI6, NDRE, RVI12 and RVI13 with CHL were all higher than 0.80. Nitrogen application level increased the sensitivity of red-edge parameter to CHL. At single nitrogen application level, the coefficient of determination between best red edge parameter and CHL was between 0.75 and 0.81. At N1 and N2 conditions the best red-edge parameter is NDDRmid, and at N3 condition the best red-edge parameter is RIDRfd (R²=0.81). The sensitivity of the red edge parameters NDDRmid, RVI5, RVI12 and DIDA to CHL at different growth stages and nitrogen application levels were all in the best 10 red-edge parameters. In the four single growth stages, the best accuracy of the random forest model was achieved, when 10 to 30 relative importance red-edge parameters were used as inputs, with R² between 0.40 and 0.89 and RMSE between 3.07 and 5.29. At four individual growth stage, the best accuracy of the model was achieved at jointing stage when 10 red edge parameters were used. At the three nitrogen application levels, the model performed best when 10 to 30 relatively important red-edge parameters were used, with R² between 0.78 and 0.87 and RMSE between 2.79 and 4.47. With the increase of nitrogen application level, the model accuracy was enhanced. At N3 condition, the model performed best when 30 relatively important red edge parameters were used with R² = 0.87 and RMSE=2.79. Step by step selection of relative importance red edge parameter features as input to optimize the random forest machine learning model improved the estimation accuracy of CHL. The best estimation accuracy in the whole growth period was R² =0.80 and RMSE=4.25. With the increase of nitrogen application level, the estimation accuracy of the model was improved, R² =0.87 and RMSE=2.79 at the condition of N3. At different growth stages and nitrogen application levels, the red-edge parameters DIDA and RVI13 were used as important features to construct the best model. The results revealed the potential of spectral red edge parameter in estimating CHL of winter wheat under different growth stages and nitrogen application conditions, and also provided a reference for the detection of chlorophyll content of other crops based on the characteristics of red edge parameter.

Leaf nitrogen content is closely related to leaf photosynthesis and the nutritional status of winter wheat plants, which directly affects the plant growth and development. While the stem nitrogen content is closely related to the proportion and content of cellulose, hemicellulose and lignin in stems, which directly affects stem quality and plant lodging resistance. However, it is still lacking on the direct estimation of stem N content in winter wheat. It is very necessary to evaluate the stalk quality and predict lodging from the perspective of stem nitrogen content. In this study, a 2-year field experiment was conducted to accurately estimate the nitrogen content in different plant organs (leaves and stems) of winter wheat. Winter wheat canopy spectral reflectance, leaf and stem nitrogen content, and leaf SPAD (Soil and Plant Analyzer Development) values were obtained at four growth stages (jointing, heading, anthesis and filling) and three nitrogen application levels (N1, N2 and N3). A systematic analysis was made to determine the sensitivity of hyperspectral vegetation indices to leaf and stem nitrogen contents at different growth stages and nitrogen application levels. Five commonly-used machine learning algorithms were used to estimate the leaf and stem nitrogen contents of winter wheat, including random forest regression (RFR), support vector regression (SVR), partial least squares regression (SVR), partial least squares regression (PLSR), Gaussian process regression (GPR) and deep neural networks (DNN). The hyperspectral vegetation indices only or combined with SPAD were used as the inputs to construct the nitrogen estimation models for leaves and stems. The results showed that the sensitivity of hyperspectral vegetation indices to the nitrogen content in leaves and stems was influenced by the growth stage and nitrogen application level. In the filling stage, the best vegetation index DCNI (Double-peak canopy nitrogen index) shared the highest sensitivity to leaf nitrogen content, where the determination coefficient R² was 0.866. The sensitivity to stem nitrogen content was the highest at the heading stage, and the R² between the best vegetation index NPQI (Normalized phaeophytinization index) and nitrogen content was 0.677. The sensitivity of the spectral vegetation index to the stem nitrogen content increased with the increasing nitrogen application level. The machine learning combined with the SPAD value and vegetation indices was improved the estimation accuracy of the nitrogen content, compared with only the vegetation index. In leaf nitrogen content, the estimation accuracy increased by 1%-7% under different growth stages and nitrogen application levels. The normalized root mean square error (NRMSE) reduced from 0.254 to 0.214 during the whole growth stage. In a single growth period, the NRMSE reduced from 0.201 to 0.128 at the heading stage, indicating the most increase. In the stem nitrogen content, the NRMSE reduced from 0.443 to 0.400 during the whole growth stage. There was the most increase during the heading stage with the values ranging from 0.323 to 0.268. In the whole growth period, the DNN model combined with the SPAD value was achieved the best accuracy to estimate the nitrogen content of leaves (R²=0.782 and NRMSE=0.214) and stems (R²=0.802 and NRMSE=0.400). The combination of the SPAD value and spectral vegetation index can be expected to improve the accuracy of the nitrogen content in the leaves and stems of winter wheat at different growth stages and nitrogen application levels.

Chlorophyll is one of the most important photosynthesis pigments in crops. The photosynthetic capacity of the plant can also indicate the plant's growth and nutritional status. Leaf chlorophyll content can be accurately acquired to monitor crop growth and yield. The portable optical instrument SPAD-502 can be used to rapidly acquire the SPAD value in a non-destructive way, indicating the leaf's relative chlorophyll content. However, the tremendous amount of hand labor cannot fully meet the needs to estimate the leaf SPAD value in a large area and then monitor the dynamic of crop growth and agricultural management. Alternatively, optical remote sensing can be expected to non-destructively measure the leaf SPAD value at a large scale. This study aims to evaluate the effects of spectral resolution on the optical SPAD estimation of winter wheat leaf using remote sensing. A four-year field experiment was carried out under four growth stages (jointing, heading, anthesis, and filling) and three levels of nitrogen application. A systematic investigation was implemented to determine the canopy spectral reflectance and leaf SPAD values of winter wheat. The leaf SPAD estimation model was constructed using 25 commonly used chlorophyll content-sensitive spectral indices combined with machine learning. An evaluation was also made on the effects of five spectral resolutions on the reflectance of a single band, spectral indices, and the estimation of SPAD value using machine learning. The results show that the sensitivity of single-band reflectance to SPAD was dominated by the spectral resolution, growth stages, and nitrogen application levels. In the whole growth period, the reflectance of the red band was more sensitive to the SPAD than the rest bands, where the determination coefficient R² was between 0.411 and 0.579 at the five spectral resolutions. The reason was that there was a strong absorption of chlorophyll in the red band. Specifically, the R² was between 0.242 and 0.700, and the Var was between 0.313 and 0.952 at the most sensitive spectral resolution for the red band in each growth stage. The red edge band reflectance at 710 nm shared the largest variation coefficient of spectral resolution sensitivity, and Var was 1.000 in the whole growth period. The main reason was that the reflectance of green vegetation changed sharply in the red edge band, indicating outstanding differences in the reflectance at different spectral resolutions. The spectral resolution also dominated the sensitivity of spectral indices to SPAD at different growth stages and nitrogen application levels. Except for the filling and anthesis stage, the spectral resolution range of the optimal sensitivity spectral index was between 25 and 50 nm. The potential reason was that the broad spectral resolution caused the spectral index to contain more spectral information, and then reduce the influence of noises, particularly for the optimal sensitivity spectral index. However, there were limited effects of spectral resolution on the sensitivity of the optimal spectral index to SPAD, compared with the single band reflectance. In the whole growth period, the spectral index mND705 presented the strongest sensitivity to SPAD at 50 nm (R² = 0.685) and a low variation coefficient of spectral resolution sensitivity (Var = 0.014). In each growth stage, the R² of the most sensitive spectral index to SPAD was between 0.387 and 0.895, and the coefficient of sensitivity variance Var was between 0.01 and 0.05. The nitrogen application level enhanced the leaf chlorophyll content to improve the sensitivity of the spectral index to detect SPAD. Spectral resolution of spectral indices was optimized as the feature inputs for machine learning models, in order to improve the estimation accuracy of SPAD. In the whole growth period, the best accuracy of estimation was achieved in the model with 25 nm spectral resolution spectral index and PLSR, indicating the R² of 0.816 and RMSE of 4.04. In each growth period, the model with the optimized spectral resolution spectral index as the input also shared the best estimation accuracy with R² between 0.531 and 0.916 and the RMSE between 2.76 and 4.47. At the three levels of nitrogen application, the R² was between 0.780 and 0.884 and the RMSE was between 2.67 and 4.55, according to the model with the optimized spectral resolution. The nitrogen application level improved the chlorophyll content of leaves. Thus the detection of SPAD was enhanced in the machine learning model using spectral characteristics. The spectral index and the spectral resolution were optimized to improve the estimation accuracy of winter wheat leaf SPAD using optical remote sensing. The finding can provide theoretical support to design the optical remote sensing sensor.