Impact of a Hyperspectral Satellite Cross-Calibration Radiometer’s Spatial and Noise Characteristics on Cross-Calibration

The satellite cross-calibration radiometer (SCR) is a conceptual on-orbit hyperspectral imaging radiometer that transfers the radiometric calibration from a “gold-standard” reference instrument such as the Landsat 8/9 Operational Land Imager (OLI) to other civil, international, or commercial “client” multispectral satellite systems via near-simultaneous cross-calibration acquisitions. The spectral resolution, spectral range, spatial resolution, and signal-to-noise ratio (SNR) all significantly impact the complexity and cost of hyperspectral SCRs, so it is important to understand their effect on cross-calibration quality. This paper discusses the results of a trade study to quantify the effects of varying ground sample distance (GSD), number of independent samples, and instrument/scene noise on cross-calibration gain uncertainties. The trade study used a simulated SCR cross-calibration with near-simultaneous nadir overpasses (SNOs) of the Landsat 8 OLI acting as the reference instrument and the DLR Earth Sensing Imaging Spectrometer (DESIS) acting as a surrogate SCR hyperspectral instrument. Results demonstrate that cross-calibration uncertainty is only minimally affected by spatial resolution and SNR, which may allow SCR instruments to be developed at a lower cost.