Evaluating the potential of red edge position (REP) of hyperspectral remote sensing data for real time estimation of LAI & chlorophyll content of kinnow mandarin (Citrus reticulata) fruit orchards

Leaf area index (LAI) and chlorophyll content (Chl) content are important indicators of plants health and nutrition status. Hyperspectral remote sensing is useful for the real-time estimation of plants biophysical and biochemical traits. The main objective of the current study was to evaluate the potential of red edge position (REP) extracted from hyperspectral reflectance profiles to estimate the LAI and Chl content of kinnow mandarin fruit orchards in Pakistan. First, we compared methods to extract the REP from hyperspectral reflectance profiles of kinnow mandarin tree canopies. Second, the ordinary least square (OLS) regression was used to investigate relationships of the derived REP with the observed kinnow mandarin traits (LAI and Chl content). Finally, kinnow mandarin traits for non-observed locations were predicted and the validity of the calibrated models was confirmed through the validation dataset. The results showed that kinnow mandarin trees depicted higher spectral response in the red edge region (680–731 nm) with REP shifting towards longer wavelengths with higher values of LAI and Chl content. Among the four algorithms tested for extraction of REP from hyperspectral reflectance, the REP models calibrated with polynomial fitting and the linear extrapolation outperformed to predict LAI and Chl content of kinnow mandarin trees with R2 values equal to 0.93 and 0.90 and RMSE values equal to 0.156% and 10.1% respectively while REP models calibrated with maximum first derivative and polynomial interpolation were less sensitive to predict these traits in kinnow mandarin trees. We conclude that the REP extracted from ground based hyperspectral reflectance can accurately estimate the kinnow mandarin LAI and Chl content and can be effectively used to assess crop health status in a wider range to manage nutritional requirement real-time in the kinnow mandarin fruit orchards. The developed mechanism will help growers for real-time monitoring orchards at different phenological stages and maximize fruit production in a precision agriculture setting.