Sun-View-Target Geometry Effects on Spectrally-Derived Vegetative Index Estimates of Absorbed Radiation and Leaf Area

Estimating the quantity and quality of vegetation is a primary goal of remote sensing. The impact of changing sun-view-target geometry on spectral measures of vegetative amount (vegetative indices, VIs) has been a major source of concern with this pursuit. A field experiment was conducted to test the statistical significance of solar and view zenith angle effects on the regression coefficients of relationships established between the simple ratio and normalized difference vegetation indices and the biophysical parameters fraction of absorbed photosynthetically active radiation (APAR) and leaf area index (LAI). Measurements of spectral bidirectional reflectance and biophysical parameters were made in an alfalfa (Medicago Sativa, L.) field located near Mead, NE throughout the 1990 growing season. View and solar zenith angle effects on VIs were found to statistically affect (at a - 0.05) the regression coefficients for relationships established between VIs and APAR, LAI. While APAR was best estimated with nadir or near nadir-derived VIs for all solar zenith angles, no one view zenith angle at all solar zenith angles best estimated LAI from the two VIs. Despite the statistical significance of view and solar zenith angle effects on regression parameters, estimation of APAR and LAI with VIs derived from off-nadir spectral data used in nadir derived regression relationships resulted in D-indices of agreement greater than 0.94 with measured values of APAR and LAI. Predicting above ground dry weight by accumulating VIs or measuring LAI (with LAI-2000) resulted in a lower mean relative error than using measured or estimated values of accumulated APAR.