Parameter estimation and evaluation of Ross-Li and RPV models for wheat phenophases using hemispherical directional reflectance measurements

Abstract

ABSTRACT Hemispherical directional reflectance factor (HDRF) measurements of wheat crop acquired with field goniometer system are used to estimate parameters and evaluate anisotropic reflectance models. The HDRFs were sampled at 54 view angles (6 azimuthal planes × 9 view zenith angles) to cover the hemispheric span up to 60° zenith. The observations were made at 7–8 days’ interval during different phenophases of the crop. Biophysical parameters were also measured parallel to HDRF. Kernel driven Ross-Li variants and RPV (Rehman Pinty Verstrate) models parameters are generated by fitting HDRF data. RossThin-H/LiSparse-R model was fitted best for all visible-near infrared bands in early phenophases while the RossThick-H/LiSparse-R model was best fitted for heading and flowering phases in most cases. In late grain development phases (Late milking and Dough) of the wheat, HDRFs of visible bands were best fitted with RossThin-H/LiDense-R model in most instances. The HDRF simulation of wheat by RPV models is identical to Ross-Li models. Both models are adequate to account reflectance anisotropy of wheat canopy. Phenophase specific model parameters can be used to reconstruct reflectance anisotropy of wheat for different applications.