Selecting the Optimum Narrow-Band Indices for Estimation of Vegetation Water Content, Using, Hyperspectral Data: Effects of Soil Background and Canopy Density

Developments in the field of hyperspectral remote sensing have provided the possibility of having new indices for estimation of vegetation biochemical and biophysical properties. The objective of this study was first to explore sensitive spectral bands that are most suitable for estimation of vegetation water content, and then investigating if soil type and canopy density affect on selecting the best narrow band index and the optimum bands for them in estimation of vegetation water content. The spectral measurements have been carried out by using a GER spectroradiometer. Two groups of narrow band vegetation indices, namely ratio based and soil based ones were compared for estimating the vegetation water. All two band combinations involving 584 wavelengths between 400 and 2400 nm were used for calculation of narrow band vegetation indices and estimating vegetation water content by using linear regression model. The predictive performances of hyperspectral indices were then determined and compared, using cross validated R2 and RMSE between the measured and estimated water content. Because of variation in canopy structure of vegetation under study and different soil backgrounds, these effects have been investigated in selecting the best narrow band indices by dividing the samples into two groups with different soil type (dark/light) and thin /thick canopy. The results indicated the better performance of all narrow band indices in light soil type and thick canopy groups, in compare with dark soil type and thin canopy groups respectively (R2CV_dark=0.85, R2CV_light=0.89, RMSECV-dark=0.35, RMSECV-light=0.29, R2CV_thin=0.78, R2CV_thick=0.81, RMSECV-thin=0.36, RMSECV-thick=0.28). The result obtained in this research highlighted the role of background effect and canopy volume in selecting the best vegetation index and optimum spectral bands.