Abstract
Monitoring the vertical profile of leaf chlorophyll (Chl) content within winter wheat canopies is of significant importance for revealing the real nutritional status of the crop. Information on the vertical profile of Chl content is not accessible to nadir-viewing remote or proximal sensing. Off-nadir or multi-angle sensing would provide effective means to detect leaf Chl content in different vertical layers. However, adequate information on the selection of sensitive spectral bands and spectral index formulas for vertical leaf Chl content estimation is not yet available. In this study, all possible two-band and three-band combinations over spectral bands in normalized difference vegetation index (NDVI)-, simple ratio (SR)- and chlorophyll index (CI)-like types of indices at different viewing angles were calculated and assessed for their capability of estimating leaf Chl for three vertical layers of wheat canopies. The vertical profiles of Chl showed top-down declining trends and the patterns of band combinations sensitive to leaf Chl content varied among different vertical layers. Results indicated that the combinations of green band (520 nm) with NIR bands were efficient in estimating upper leaf Chl content, whereas the red edge (695 nm) paired with NIR bands were dominant in quantifying leaf Chl in the lower layers. Correlations between published spectral indices and all NDVI-, SR- and CI-like types of indices and vertical distribution of Chl content showed that reflectance measured from 50°, 30° and 20° backscattering viewing angles were the most promising to obtain information on leaf Chl in the upper-, middle-, and bottom-layer, respectively. Three types of optimized spectral indices improved the accuracy for vertical leaf Chl content estimation. The optimized three-band CI-like index performed the best in the estimation of vertical distribution of leaf Chl content, with R2 of 0.84–0.69, and RMSE of 5.37–5.56 µg/cm2 from the top to the bottom layers, while the optimized SR-like index was recommended for the bottom Chl estimation due to its simple and universal form. We suggest that it is necessary to take into account the penetration characteristic of the light inside the canopy for different Chl absorption regions of the spectrum and the formula used to derive spectral index when estimating the vertical profile of leaf Chl content using off-nadir hyperspectral data.
Highlights
Dynamic monitoring of chlorophyll (Chl) content of crop plants during different growth stages provides crucial information to understand plants’ physiological status and their response to environmental changes in agro-ecosystems [1,2]
Based on the analyses presented above, we calculated coefficients of determination for the relationships between vertical leaf Chl content and the chlorophyll index (CI)-like indices calculated from all possible three-band combinations in the regions of the shorter green to red edge bands (470–730 nm), the red to NIR bands (680–800 nm) and the NIR bands (740–1000 nm) at the nadir and backscattering observations
Our results show the importance of accounting for the sensitivity of spectral bands and penetration characteristics of the light in the selected spectral region inside the canopy, as well as the formula that was used to derive the spectral index when estimating vertical profile of leaf Chl within a wheat canopy
Summary
Dynamic monitoring of chlorophyll (Chl) content of crop plants during different growth stages provides crucial information to understand plants’ physiological status and their response to environmental changes in agro-ecosystems [1,2]. Crop canopies generally exhibit vertical heterogeneity of leaf Chl because of different light environment, growth and development of the plant itself, nitrogen uptake and remobilization processes [5]. Since the lower leaves of the canopy are more prone to nutrient deficits and plant senescence than the upper leaves [8], knowledge of leaf Chl content in the lower layers has significant interest, e.g., in precision agriculture. It would allow the early detection of subtle variations within the field, rather than taking steps after crop stress has happened
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
