An Empirical Comparison of Calibration and Validation Methodologies for Airborne Imaging Spectroscopy

To date, a large number of existing applications in India have used multi-band observations from airborne and spaceborne platforms. New sensors are providing additional capabilities thanks to special aerial missions with the compact airborne spectrographic imager (CASI), the short-wave infrared (SWIR) full spectrum imager (SFSI) and the National Aeronautics and Space Administration’s (NASA’s) Next Generation Airborne Visible/Infrared Imaging Spectrometer (AVIRIS-NG). Opportunities to exploit quantitative spectroscopic signatures and high spatial resolution have garnered great interest among the scientific community, and the success of these missions will rely on accurate calibration. Here we focus on a vicarious calibration experiment conducted for the AVIRIS-NG India campaign. We discuss initial validation results, with descriptions of in situ and remote calibration and measurement protocols, geometric processing with precise position and attitude data, and atmospheric simulations used to validate the remote measurement. A partnership between Indian Space Research Organisation (ISRO) and NASA investigators proved a unique opportunity to assess the empirical variability in results, indicating their sensitivity to modelling choices and assumptions. The vicarious calibration exercise uses multiple radiative transfer models, including MODTRAN 6.0 and a new version of the 6S radiative transfer code, viz. 6SV2.1, which is capable of accounting for polarization.