The sensitivity of the OSAVI vegetation index to observational parameters

摘要

OSAVI is a soil-adjusted vegetation index optimized for agricultural monitoring. The SAIL-SOILSPECT-PROSPECT-KUUSK model was applied to test the sensitivity of OSAVI to a number of observational and environmental parameters, with the constraint that the parameter ranges were chosen to be representative of conditions encountered during operational monitoring by SPOT and Landsat of the U.K. sugar beet crop in the summer months. Within these defined limits, variations in solar and viewing angles introduced no more than 4% systematic (and rectifiable)error in estimation of crop canopy cover while errors associated with atmospheric correction were of a similar magnitude. Although sugar beet has large leaves close to the ground and was expected to generate a large hot spot effect, the effect of ;uncertainty in leaf size appears to be negligible. Leaf angle distribution is effectively accounted for by the use of crop cover as an index of productivity, except when the crop is wilting, but this is unlikely to occur in sugar beet at the time of satellite overpass. Subject to the uncertainties of modeling, the estimation of crop canopy cover within an absolute error of 5% appears to be achievable. (C) Elsevier Science Inc., 1998. [References: 16]