Evaluation of BIAS reduction in cross-calibration of NDVI based on soil isoline equations: Comparison with error estimated from signal-to-noise ratio

Intercalibration among remotely-sensed data products such as spectral vegetation indices (VIs) from sensors onboard different satellites has been investigated for integrational use of multiple datasets. To facilitate intercalibration technique of VIs, the relationships among the VI values from different sensors need to be fully understood. This study evaluated the accuracies of NDVI translation technique based on soil iso-line equations. The evaluation was performed by comparing the errors in intersensor relationships of NDVI simulated by radiative transfer model and propagted error in NDVI from the sensor specific signal-to-noise ratio (SNR) of each band. The numerical results indicated that the translation equation showed enough accuracy relative to the propagated error originated from SNR, if the soil brightness underneath the canopy layer can be estimated prior to the translation.