Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Crop Production Systems Research » Research » Publications at this Location » Publication #315926
Title: Quantitative estimation of the fluorescent parameters for crop leaves with the Bayesian inversion

Author
item ZHAO, FENG - Beihang University
item GUO, YIQING - Beihang University
item Huang, Yanbo
item ZHAO, HUIJIE - Beihang University
item LIU, GUANG - Beihang University

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/22/2015
Publication Date: 10/27/2015
Citation: Zhao, F., Guo, Y., Huang, Y., Zhao, H., Liu, G. 2015. Quantitative estimation of the fluorescent parameters for crop leaves with the Bayesian inversion. Remote Sensing. 7:14179-14199.

Interpretive Summary: Monitoring crop growth and stress conditions is necessary for crop planning and management. Remote sensing is effective for rapid crop monitoring. Scientists of Beihang University, Beijing, China and USDA-ARS Crop Production Systems Research Unit, Stoneville, Mississippi, collaborated to estimate the fluorescent parameters of soybean and cotton leaves from the leaf hyperspectral measurements through inverting the FluorMODleaf model, a leaf-level fluorescence model as part of a widely-used leaf optical model. The results indicated that the innovative Bayesian inversion approach can be used to extract the fluorescent parameters of plant leaves by inverting the FluorMODleaf model. With the inversion of the FluorMODleaf model, the fluorescence signals can be more effectively used to monitor crop growing status and stress conditions.

Technical Abstract: In this study, the fluorescent parameters of crop leaves were retrieved from the leaf hyperspectral measurements by inverting the FluorMODleaf model, which is a leaf-level fluorescence model that is based on the widely used and validated PROSPECT (leaf optical properties) model and can simulate the chlorophyll fluorescence spectra for both sides of the leaves. Firstly, a sensitivity analysis of the FluorMODleaf model was performed using the EFAST (Extended Fourier Amplitude Sensitivity Test) method. It showed that the fluorescence lifetimes of photosystem I (PSI) and photosystem II (PSII) and the leaf chlorophyll contents are the most sensitive parameters among all eight inputs of the FluorMODleaf model. Based on the results of sensitivity analysis, the FluorMODleaf model was inverted using the leaf fluorescence spectra measured for soybean and cotton leaves. In order to achieve stable inversion results, Bayesian inference theory was applied in the inversion process. The relative absorption cross section of PSI and PSII (d) and the fluorescence lifetimes of PSI and PSII (tI and tII) of the FluorMODleaf model were retrieved with the Bayesian inversion approach. Results showed that coefficient of determination (R2) between the fluorescence spectra reconstructed from the inverted fluorescent parameters and the fluorescence spectra measured in the experiment is 0.9566 for soybean and 0.9359 for cotton, and the root mean square error (RMSE) is 4.2 W·m-2·sr-1·nm-1 for soybean and 3.4 W·m-2·sr-1·nm-1 for cotton. The inverted values of the fluorescent parameters were within the reasonable range, which indicates a reliable inversion result. Based on the results, it can be concluded that the Bayesian inversion approach can be used to retrieve the fluorescent parameters of plant leaves by inverting the FluorMODleaf model. By inverting the FluorMODleaf model, researcher can more effectively use the fluorescence signals to monitor the growing status of plants and the stress conditions of crops. Further works should be on further improving the retrieval accuracy of the fluorescent parameters of plant leaves and investigating the potential of this method.