Vegetation canopy PAR absorptance and the normalized difference vegetation index: An assessment using the SAIL model

Abstract

The SAIL canopy radiative transfer model is employed to examine relations between the normalized difference vegetation index (NDVI) and absorbed photosynthetically active radiation (APAR) in a herbaceous vegetation canopy. This analysis extends previous work by using a bidirectional canopy radiative transfer model to consider relations between instantaneous NDVI measurements and diurnally integrated canopy APAR capacity. Effective use of periodic, instantaneous remote sensing observations in landscape process studies depends on stable estimation of diurnally integrated conditions. The results from this analysis show that instantaneous NDVI measurements provide a stable and near-linear estimate of vegetation canopy diurnal APAR capacity, as long as the remotely sensed observations are taken when the solar zenith angle is above 60° and the sensor views within 40° from nadir. The direct / diffuse ratio of incident radiance produces little variance in the relation. Variations in canopy optical properties and architecture, as well as background spectral reflectance, also cause deviations in the APAR / NDVI relation, with background spectral reflectance properties producing the largest effect. The practical significance of this landscape variability in remotely sensed APAR assessment is uncertain. Remote sensing observation strategies are explored which might be used to evaluate and ultimately overcome the influence of landscape variance in APAR assessment. The capacity of periodic, instantaneous remotely sensed observations to characterize the diurnally integrated APAR in vegetation canopies appears reasonable within the constraints noted.