The sensitivity based estimation of leaf area index from spectral vegetation indices

The performances of seven spectral vegetation indices (SVIs) were investigated for their sensitivity to a varying range of LAI. The evaluation was carried out for a dataset collected using SPOT 5 HRG 10m imagery and simulated spectra using PROSPECT+SAIL reflectance models with varying soil reflectance backgrounds. The aim was to evaluate the applicability of multiple SVIs for LAI mapping based on the sensitivity analysis. The main sensitivity function was the first derivative of the regression function divided by the standard errors of the SVIs. In addition, the sensitivity of individual band and SVI with LAI was carried out using the ordinary least squares regressions. A new SVI, reduced infrared simple ratio (RISR) was developed based on an empirical red modification to infrared simple ratio (ISR) SVI. The new SVI was demonstrated which has significantly reduced the effect of soil background reflectance while maintaining high sensitivity to a wide range of LAI.