Using spaceborne hyperspectral data for spectral cross-calibration of multispectral sensors

Within this paper, a methodology for the cross-calibration of multispectral sensors based on spaceborne hyperspectral, and its practical implementation for a consistent re-processing of 30+ years of NOAA AVHRR data are shown. For the development, a large variety of HYPERION images including typical surface spectral signatures for Western, Central, South and Eastern Europe, Scandinavia, as well as Northern Africa are used, also covering different seasonal and phenological conditions. This extensive database of hyperspectral imagery is then used to simulate the spectral responses of various multispectral sensors. Extending the approach from STEVEN et al. [1, 2], statistically robust models are developed in order to derive intercalibration factors for the specific sensor spectral responses. By applying these factors, differences caused by the spectral response functions of multispectral sensors can be reduced resulting in an increased consistency in multi-sensoral time series. This finding underpins the need for spaceborne imaging spectrometers such as HYPERION, the upcoming EnMAP and HISUI, and dedicated missions such as the proposed TRUTHS, which can be used as a direct reference for spectral and radiometric cross-calibration.